| aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ALCALA_AVW | 70.9 | 39.0 | 43.1 | 53.0 | 14.7 | 56.3 | 38.2 | 47.7 | 40.2 | 28.2 | 29.2 | 34.5 | 39.1 | 39.7 | 68.5 | 18.5 | 33.7 | 34.7 | 65.2 | 44.4 |
| ALCALA_LAVW | 70.4 | 34.8 | 43.4 | 49.2 | 16.6 | 57.3 | 39.7 | 45.6 | 39.7 | 26.7 | 22.6 | 30.3 | 41.0 | 39.6 | 67.5 | 18.6 | 32.5 | 27.3 | 64.9 | 42.6 |
| CASIA_SVM-MULTIFEAT | 78.2 | 45.8 | 48.7 | 50.7 | 27.5 | 61.1 | 47.4 | 49.2 | 52.2 | 30.5 | 37.1 | 42.0 | 52.2 | 51.9 | 81.6 | 28.7 | 29.7 | 34.6 | 67.9 | 53.0 |
| CNRS_FUSE-KNN-CTS | 66.8 | 33.2 | 35.8 | 45.8 | 20.0 | 49.0 | 38.5 | 35.7 | 41.7 | 22.8 | 23.4 | 32.2 | 35.3 | 35.5 | 67.5 | 16.6 | 26.7 | 20.0 | 55.3 | 40.6 |
| CNRS_KNN-CT | 57.6 | 19.8 | 31.6 | 40.4 | 17.1 | 40.1 | 35.7 | 32.1 | 35.6 | 21.8 | 18.1 | 26.8 | 22.5 | 29.1 | 65.3 | 17.0 | 16.6 | 16.4 | 41.4 | 33.6 |
| CNRS_KNN-GABOR | 51.8 | 21.8 | 31.8 | 29.0 | 18.3 | 37.6 | 33.4 | 28.5 | 36.8 | 17.7 | 18.7 | 23.3 | 17.3 | 28.9 | 62.9 | 16.6 | 17.9 | 22.7 | 35.5 | 36.6 |
| CNRS_KNN-HRGB | 42.1 | 12.0 | 22.5 | 29.7 | 13.8 | 14.3 | 23.7 | 25.0 | 27.3 | 13.3 | 16.1 | 23.5 | 14.1 | 20.4 | 60.8 | 13.2 | 17.3 | 10.9 | 18.6 | 22.7 |
| CNRS_KNN-OSHL | 61.5 | 29.4 | 26.0 | 38.4 | 17.2 | 41.3 | 25.6 | 28.6 | 34.3 | 12.9 | 20.4 | 24.8 | 31.7 | 23.4 | 61.0 | 17.5 | 24.6 | 20.6 | 43.6 | 36.3 |
| CVC_FLAT | 85.3 | 57.8 | 66.0 | 66.1 | 36.2 | 70.6 | 60.6 | 63.5 | 55.1 | 44.6 | 53.4 | 49.1 | 64.4 | 66.8 | 84.8 | 37.4 | 44.1 | 47.9 | 81.9 | 67.5 |
| CVC_FLAT-HOG-ESS | 86.3 | 60.7 | 66.4 | 65.3 | 41.0 | 71.7 | 64.7 | 63.9 | 55.5 | 40.1 | 51.3 | 45.9 | 65.2 | 68.9 | 85.0 | 40.8 | 49.0 | 49.1 | 81.8 | 68.6 |
| CVC_PLUS | 86.6 | 58.4 | 66.7 | 67.3 | 34.8 | 70.4 | 60.0 | 64.2 | 52.5 | 43.0 | 50.8 | 46.5 | 64.1 | 66.8 | 84.4 | 37.5 | 45.1 | 45.4 | 82.1 | 67.0 |
| FIRSTNIKON_AVGSRKDA | 83.3 | 59.3 | 62.7 | 65.3 | 30.2 | 71.6 | 58.2 | 62.2 | 54.3 | 40.7 | 49.2 | 50.0 | 66.6 | 62.9 | 83.3 | 34.2 | 48.2 | 46.1 | 83.4 | 65.5 |
| FIRSTNIKON_AVGSVM | 83.8 | 58.2 | 62.6 | 65.2 | 32.0 | 69.8 | 57.7 | 61.1 | 54.5 | 44.0 | 50.3 | 49.6 | 64.6 | 61.7 | 83.2 | 33.4 | 46.5 | 48.0 | 81.6 | 65.3 |
| FIRSTNIKON_BOOSTSRKDA | 83.0 | 59.2 | 61.4 | 64.6 | 33.2 | 71.1 | 57.5 | 61.0 | 54.8 | 40.7 | 48.3 | 50.0 | 65.5 | 63.4 | 82.8 | 32.8 | 47.0 | 47.1 | 83.3 | 64.6 |
| FIRSTNIKON_BOOSTSVMS | 83.5 | 56.8 | 61.8 | 65.5 | 33.2 | 69.7 | 57.3 | 60.5 | 54.6 | 43.1 | 48.3 | 50.3 | 64.3 | 62.4 | 82.3 | 32.9 | 46.9 | 48.4 | 82.0 | 64.2 |
| FIRST_L2MKL | 6.8 | 6.9 | 16.8 | 6.7 | 8.9 | 4.1 | 12.4 | 9.0 | 13.7 | 5.8 | 4.7 | 18.5 | 5.9 | 5.4 | 49.5 | 6.1 | 3.4 | 8.0 | 13.8 | 5.6 |
| HAS_FISHSIFT-FISHSEG | 51.0 | 26.1 | 24.0 | 30.4 | 11.6 | 32.4 | 21.6 | 34.6 | 23.1 | 11.3 | 9.1 | 31.2 | 15.6 | 18.4 | 63.2 | 9.4 | 13.1 | 10.9 | 26.6 | 25.3 |
| IIR_SVM-ROI-IC | 74.6 | 49.9 | 45.1 | 48.6 | 19.6 | 59.6 | 46.2 | 47.8 | 47.3 | 33.8 | 36.0 | 36.8 | 51.9 | 50.0 | 73.4 | 15.7 | 37.2 | 36.5 | 69.9 | 54.0 |
| KERLE_SVM-DENSESIFT | 74.2 | 45.6 | 46.5 | 55.6 | 20.5 | 59.7 | 48.7 | 48.2 | 45.9 | 23.0 | 31.5 | 41.2 | 50.6 | 49.8 | 76.5 | 22.0 | 30.8 | 41.7 | 66.2 | 50.4 |
| LEAR_CHI-SVM-MULT | 79.7 | 51.8 | 54.3 | 61.3 | 30.8 | 66.7 | 52.0 | 55.7 | 52.6 | 38.7 | 42.0 | 45.0 | 58.0 | 61.2 | 81.0 | 27.6 | 36.3 | 41.5 | 73.8 | 58.9 |
| LEAR_CHI-SVM-MULT-LOC | 79.5 | 55.5 | 54.5 | 63.9 | 43.7 | 70.3 | 66.4 | 56.5 | 54.4 | 38.8 | 44.1 | 46.2 | 58.5 | 64.2 | 82.2 | 39.1 | 41.3 | 39.8 | 73.6 | 66.2 |
| LEOBEN_DENSESIFT | 77.0 | 48.5 | 53.3 | 57.0 | 28.9 | 63.9 | 51.9 | 52.1 | 48.5 | 30.8 | 31.1 | 43.6 | 54.7 | 55.5 | 77.2 | 19.8 | 36.0 | 46.3 | 71.7 | 54.9 |
| LEOBEN_SCC-200 | 80.4 | 49.6 | 54.9 | 60.7 | 23.6 | 64.6 | 54.0 | 52.5 | 50.8 | 31.1 | 43.2 | 43.1 | 55.8 | 56.7 | 80.4 | 29.7 | 41.7 | 43.0 | 73.8 | 58.8 |
| LEOBEN_SCC-CLS | 79.5 | 52.1 | 57.2 | 59.9 | 29.3 | 63.5 | 55.1 | 53.9 | 51.1 | 31.3 | 42.9 | 44.1 | 54.8 | 58.4 | 81.1 | 30.0 | 40.2 | 44.2 | 74.9 | 58.2 |
| LIG_MIRIM-VPH | 62.0 | 29.1 | 26.9 | 29.6 | 12.2 | 26.0 | 23.5 | 33.6 | 35.7 | 14.8 | 10.6 | 22.2 | 20.5 | 18.5 | 64.6 | 10.1 | 16.9 | 11.3 | 20.7 | 36.0 |
| LIG_MRIM-COLORSIFT | 69.5 | 37.0 | 38.8 | 40.4 | 23.1 | 48.2 | 36.3 | 40.6 | 41.9 | 17.4 | 31.5 | 29.2 | 39.1 | 35.2 | 72.9 | 23.1 | 29.3 | 22.9 | 52.8 | 40.1 |
| LIG_MRIM-FUSION | 71.6 | 41.2 | 40.6 | 45.5 | 25.1 | 54.6 | 39.5 | 43.9 | 46.6 | 19.7 | 32.3 | 33.9 | 44.3 | 43.0 | 74.3 | 23.3 | 31.0 | 24.8 | 60.8 | 43.0 |
| LIP6_HB-SPK-SVM | 77.9 | 49.8 | 49.5 | 56.7 | 27.2 | 63.0 | 51.4 | 52.9 | 49.8 | 33.2 | 37.7 | 42.8 | 56.4 | 53.8 | 77.7 | 21.7 | 36.3 | 39.5 | 72.7 | 57.7 |
| LIP6_SS-SPK-SVM | 80.9 | 52.3 | 53.8 | 60.8 | 29.1 | 66.2 | 53.4 | 55.9 | 50.7 | 33.8 | 43.9 | 44.6 | 59.4 | 58.0 | 80.0 | 25.3 | 41.9 | 42.5 | 78.4 | 60.1 |
| LIRIS_BASELINE | 73.5 | 44.3 | 46.0 | 53.7 | 23.9 | 55.5 | 47.2 | 43.9 | 47.1 | 18.3 | 35.5 | 37.0 | 47.3 | 44.5 | 76.7 | 24.6 | 32.6 | 35.4 | 64.8 | 48.7 |
| LIRIS_EER | 74.1 | 44.0 | 45.4 | 54.9 | 23.5 | 56.8 | 46.9 | 43.7 | 47.1 | 18.3 | 35.4 | 37.4 | 46.9 | 44.5 | 76.7 | 24.6 | 30.8 | 35.7 | 63.6 | 48.8 |
| LIRIS_SOFT-BASELINE | 70.0 | 33.8 | 40.6 | 47.3 | 20.7 | 50.0 | 42.8 | 38.5 | 43.7 | 19.9 | 32.7 | 34.1 | 36.7 | 34.8 | 73.0 | 23.0 | 28.7 | 24.1 | 57.5 | 41.1 |
| LIRIS_SOFT-EER | 70.3 | 33.7 | 41.4 | 48.6 | 21.0 | 51.1 | 42.8 | 38.3 | 43.6 | 20.1 | 32.9 | 34.4 | 36.2 | 34.7 | 73.2 | 23.1 | 25.6 | 24.9 | 57.2 | 40.5 |
| MPI_STRUCT | 75.9 | 49.3 | 44.4 | 48.7 | 24.3 | 66.3 | 50.3 | 52.7 | 37.0 | 35.4 | 38.5 | 43.1 | 55.2 | 62.1 | 67.9 | 22.7 | 40.5 | 44.4 | 68.4 | 49.8 |
| NECUIUC_CDCV | 88.1 | 68.0 | 68.0 | 72.5 | 41.0 | 78.9 | 70.4 | 70.4 | 58.1 | 53.4 | 55.7 | 59.3 | 73.1 | 71.3 | 84.5 | 32.3 | 53.3 | 56.7 | 86.0 | 66.8 |
| NECUIUC_CLS-DTCT | 88.0 | 68.6 | 67.9 | 72.9 | 44.2 | 79.5 | 72.5 | 70.8 | 59.5 | 53.6 | 57.5 | 59.0 | 72.6 | 72.3 | 85.3 | 36.6 | 56.9 | 57.9 | 85.9 | 68.0 |
| NECUIUC_LL-CDCV | 87.1 | 67.4 | 65.8 | 72.3 | 40.9 | 78.3 | 69.7 | 69.7 | 58.5 | 50.1 | 55.1 | 56.3 | 71.8 | 70.8 | 84.1 | 31.4 | 51.5 | 55.1 | 84.7 | 65.2 |
| NECUIUC_LN-CDCV | 87.7 | 67.8 | 68.1 | 71.1 | 39.1 | 78.5 | 70.6 | 70.7 | 57.4 | 51.7 | 53.3 | 59.2 | 71.6 | 70.6 | 84.0 | 30.9 | 51.7 | 55.9 | 85.9 | 66.7 |
| RITSU_AKF | 75.9 | 52.3 | 50.6 | 55.9 | 25.4 | 65.4 | 48.9 | 50.4 | 48.6 | 36.2 | 44.4 | 41.9 | 54.7 | 52.9 | 76.8 | 17.1 | 38.4 | 39.3 | 72.5 | 54.3 |
| RITSU_ASF | 75.4 | 51.4 | 50.4 | 55.7 | 24.4 | 65.7 | 48.7 | 49.5 | 48.8 | 32.4 | 42.7 | 41.9 | 54.1 | 52.8 | 76.1 | 17.1 | 38.6 | 39.5 | 72.3 | 53.1 |
| RITSU_WSF | 76.9 | 51.7 | 50.7 | 55.3 | 28.4 | 65.4 | 47.4 | 50.1 | 48.5 | 36.3 | 43.3 | 41.0 | 54.8 | 54.2 | 76.7 | 16.9 | 38.8 | 38.8 | 72.5 | 53.9 |
| TSINGHUA_ALL-SVM-BOOST | 45.5 | 14.5 | 29.2 | 33.9 | 21.0 | 22.1 | 25.8 | 22.4 | 29.8 | 9.7 | 18.9 | 25.0 | 20.6 | 27.3 | 65.9 | 8.4 | 20.8 | 17.7 | 32.9 | 28.3 |
| TSINGHUA_SVM-SEG-HOG | 32.7 | 6.0 | - | 15.1 | 11.1 | 9.3 | 17.4 | 8.5 | 13.4 | 2.6 | 7.9 | 12.5 | - | - | 57.8 | 9.6 | 5.3 | 6.9 | 7.9 | 9.2 |
| UC3M_GEN-DIS | 69.9 | 43.8 | 33.5 | 36.1 | 27.6 | 51.6 | 49.1 | 41.1 | 44.4 | 24.3 | 29.8 | 36.4 | 38.8 | 51.1 | 72.5 | 19.9 | 21.3 | 22.9 | 51.4 | 41.2 |
| UVASURREY_BASELINE | 84.1 | 59.2 | 62.7 | 65.4 | 35.7 | 70.6 | 59.8 | 61.3 | 56.7 | 45.3 | 52.4 | 50.6 | 66.1 | 66.6 | 83.7 | 34.8 | 47.2 | 47.7 | 80.8 | 65.9 |
| UVASURREY_MKFDA+BOW | 84.7 | 63.9 | 66.1 | 67.3 | 37.9 | 74.1 | 63.2 | 64.0 | 57.1 | 46.2 | 54.7 | 53.5 | 68.1 | 70.6 | 85.2 | 38.5 | 47.2 | 49.3 | 83.2 | 68.1 |
| UVASURREY_TUNECOLORKERNELSEL | 85.0 | 62.8 | 65.1 | 66.5 | 37.6 | 73.5 | 62.1 | 62.0 | 57.4 | 45.1 | 54.5 | 52.5 | 67.7 | 69.8 | 84.8 | 39.1 | 46.8 | 49.9 | 82.9 | 68.1 |
| UVASURREY_TUNECOLORSPECKDA | 84.6 | 62.4 | 65.6 | 67.2 | 39.4 | 74.0 | 63.4 | 62.8 | 56.7 | 43.8 | 54.7 | 52.7 | 67.3 | 70.6 | 85.0 | 38.8 | 46.9 | 50.0 | 82.2 | 66.2 |
| aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|---|
| aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CASIA_SVM-PHOG | 19.0 | 15.4 | 9.7 | 9.5 | - | 21.0 | - | - | 2.8 | - | 2.4 | - | - | - | - | - | - | - | 16.1 | - |
| CASIA_SVM-PHOG+COLOR | 26.7 | 20.5 | 10.2 | 10.2 | 9.5 | 26.6 | 13.3 | 12.7 | 9.5 | 7.6 | 10.2 | 11.1 | 16.6 | 22.1 | 15.8 | 9.4 | 4.2 | 10.1 | 25.3 | 16.1 |
| CVC_HOG-BOW | 35.8 | 27.6 | 10.2 | 10.1 | 17.2 | 32.1 | 21.0 | 18.9 | 13.0 | 10.9 | 17.1 | 14.2 | 24.5 | 28.8 | 18.0 | 10.3 | 16.0 | 13.1 | 25.9 | 27.3 |
| CVC_HOG-BOW-ESS-FLAT | 35.5 | 27.5 | 11.1 | 11.2 | 16.7 | 32.2 | 20.8 | 19.2 | 13.9 | 14.6 | 16.3 | 12.1 | 29.0 | 29.0 | 18.8 | 11.6 | 18.4 | 19.4 | 30.6 | 26.6 |
| LEAR_CHI-SVM-SIFT-HOG | 27.1 | 30.2 | 9.8 | 10.7 | 19.6 | 36.0 | 32.1 | 12.5 | 11.2 | 14.0 | 16.4 | 10.2 | 22.6 | 27.8 | 19.9 | 11.6 | 16.5 | 11.9 | 34.5 | 32.1 |
| LEAR_CHI-SVM-SIFT-HOG-CLS | 28.4 | 30.7 | 11.0 | 12.4 | 21.4 | 36.2 | 32.2 | 14.1 | 12.0 | 18.5 | 17.8 | 15.6 | 25.7 | 29.5 | 20.5 | 12.8 | 20.8 | 14.2 | 35.1 | 34.7 |
| MIZZOU_DEF-HOG-LBP | 11.4 | 27.5 | 6.0 | 11.1 | 27.0 | 38.8 | 33.7 | 25.2 | 15.0 | 14.4 | 16.9 | 15.1 | 36.3 | 40.9 | 37.0 | 13.2 | 22.8 | 9.6 | 3.5 | 32.1 |
| MIZZOU_DEF-HOG-LBP-WOCONTEXT | 25.0 | 27.9 | 6.1 | 10.2 | 26.6 | 38.0 | 33.9 | 21.9 | 14.5 | 17.5 | 16.8 | 17.0 | 35.3 | 40.0 | 36.6 | 11.7 | 22.3 | 15.6 | 33.6 | 32.7 |
| MPI_STRUCT | 41.0 | 22.4 | 10.6 | 12.0 | 9.1 | 30.2 | 12.9 | 31.1 | 4.5 | 13.7 | 15.0 | 21.2 | 21.3 | 29.9 | 11.6 | 9.1 | 10.5 | 22.4 | 30.3 | 11.3 |
| NECUIUC_CLS-DTCT | 44.9 | 33.1 | 12.3 | 10.5 | 11.0 | 43.4 | 28.4 | 30.9 | 11.1 | 20.1 | 22.9 | 25.1 | 33.7 | 38.2 | 22.5 | 11.0 | 22.9 | 23.4 | 32.1 | 24.8 |
| OXFORD_MKL | 47.8 | 39.8 | 17.4 | 15.8 | 21.9 | 42.9 | 27.7 | 30.5 | 14.6 | 20.6 | 22.3 | 17.0 | 34.6 | 43.7 | 21.6 | 10.2 | 25.1 | 16.6 | 46.3 | 37.6 |
| TSINGHUA_SVM-SEG-HOG | 9.1 | - | - | 2.3 | 9.1 | - | 9.1 | - | - | 0.0 | - | 0.4 | - | 9.1 | 1.2 | 0.0 | 0.0 | - | 1.1 | 0.0 |
| TTIWEIZ_NNHOUGH | 23.8 | 24.0 | - | - | - | 21.9 | 21.0 | - | - | 14.3 | - | - | 19.6 | 24.0 | - | - | - | - | - | 23.2 |
| UC3M_GEN-DIS | 22.4 | 17.1 | 10.4 | 9.5 | 9.1 | 18.6 | 11.0 | 22.0 | 9.2 | 10.0 | 10.5 | 16.5 | 15.1 | 21.8 | 11.5 | 9.2 | 9.9 | 11.4 | 17.1 | 2.6 |
| UVA_BAGOFWINDOWS | 32.5 | 23.7 | 10.6 | 8.4 | 3.2 | 28.2 | 14.4 | 33.7 | 1.2 | 13.2 | 16.3 | 23.2 | 24.6 | 30.7 | 13.1 | 4.5 | 9.3 | 28.0 | 29.0 | 9.5 |
| UVA_BOWSEG | 40.2 | - | 6.9 | - | - | 26.4 | - | 34.0 | - | - | 19.0 | - | - | - | - | - | - | 21.2 | 27.2 | - |
| UoCTTI_LSVM-MDPM | 39.5 | 46.8 | 13.5 | 15.0 | 28.5 | 43.8 | 37.2 | 20.7 | 14.9 | 22.8 | 8.7 | 14.4 | 38.0 | 42.0 | 41.5 | 12.6 | 24.2 | 15.8 | 43.9 | 33.5 |
| aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BERKELEY_POSELETS | - | - | - | - | - | - | - | - | - | - | - | - | - | - | 43.2 | - | - | - | - | - |
- Entries in parentheses are synthesized from detection results.
| [mean] | back ground |
aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BONN_SVM-SEGM | 36.3 | 83.9 | 64.3 | 21.8 | 21.7 | 32.0 | 40.2 | 57.3 | 49.4 | 38.8 | 5.2 | 28.5 | 22.0 | 19.6 | 33.6 | 45.5 | 33.6 | 27.3 | 40.4 | 18.1 | 33.6 | 46.1 |
| BROOKESMSRC_AHCRF | 24.8 | 79.6 | 48.3 | 6.7 | 19.1 | 10.0 | 16.6 | 32.7 | 38.1 | 25.3 | 5.5 | 9.4 | 25.1 | 13.3 | 12.3 | 35.5 | 20.7 | 13.4 | 17.1 | 18.4 | 37.5 | 36.4 |
| (CASIA_SVM-PHOG+COLOR) | 10.3 | 24.5 | 8.2 | 4.3 | 6.5 | 4.8 | 14.7 | 27.2 | 10.0 | 7.0 | 4.2 | 6.5 | 4.9 | 3.6 | 10.3 | 14.5 | 11.2 | 8.1 | 10.6 | 4.9 | 13.3 | 17.6 |
| CVC_HOCRF | 34.5 | 80.2 | 67.1 | 26.6 | 30.3 | 31.6 | 30.0 | 44.5 | 41.6 | 25.2 | 5.9 | 27.8 | 11.0 | 23.1 | 40.5 | 53.2 | 32.0 | 22.2 | 37.4 | 23.6 | 40.3 | 30.2 |
| (CVC_HOG-BOW) | 9.4 | 2.3 | 9.4 | 6.1 | 3.4 | 5.2 | 13.5 | 21.8 | 15.2 | 6.6 | 2.2 | 7.0 | 1.6 | 5.6 | 11.1 | 11.7 | 16.8 | 3.2 | 13.8 | 7.3 | 17.3 | 16.5 |
| (CVC_HOG-BOW-ESS-FLAT) | 0.9 | 2.0 | 1.7 | 0.2 | 0.1 | 0.3 | 0.6 | 0.1 | 0.2 | 1.2 | 0.3 | 0.8 | 0.4 | 0.1 | 0.2 | 0.1 | 1.7 | 1.1 | 1.1 | 2.0 | 0.2 | 4.6 |
| (LEAR_CHI-SVM-SIFT-HOG) | 8.4 | 7.1 | 7.4 | 4.2 | 2.9 | 5.2 | 15.1 | 19.7 | 17.7 | 5.8 | 2.3 | 6.3 | 2.6 | 3.7 | 7.0 | 6.3 | 9.7 | 7.6 | 9.2 | 7.5 | 12.8 | 16.8 |
| (LEAR_CHI-SVM-SIFT-HOG-CLS) | 5.9 | 6.8 | 4.0 | 2.1 | 1.4 | 1.8 | 3.7 | 19.5 | 11.8 | 5.3 | 3.8 | 0.1 | 3.6 | 3.0 | 2.1 | 2.8 | 8.0 | 0.9 | 4.4 | 8.3 | 6.9 | 22.8 |
| LEAR_SEGDET | 25.7 | 79.1 | 44.6 | 15.5 | 20.5 | 13.3 | 28.8 | 29.3 | 35.8 | 25.4 | 4.4 | 20.3 | 1.3 | 16.4 | 28.2 | 30.0 | 24.5 | 12.2 | 31.5 | 18.3 | 28.8 | 31.9 |
| (MIZZOU_DEF-HOG-LBP) | 7.5 | 0.6 | 1.2 | 2.5 | 0.0 | 0.0 | 18.4 | 33.7 | 13.3 | 7.6 | 0.7 | 2.2 | 1.1 | 2.8 | 13.8 | 22.7 | 14.9 | 0.6 | 15.5 | 0.0 | 0.0 | 5.5 |
| (MIZZOU_DEF-HOG-LBP-WOCONTEXT) | 5.4 | 0.6 | 1.3 | 2.9 | 0.0 | 0.0 | 22.7 | 5.0 | 9.5 | 3.2 | 1.3 | 2.8 | 0.3 | 1.4 | 8.9 | 7.9 | 11.9 | 0.9 | 10.9 | 2.8 | 3.5 | 14.9 |
| MPI_A2 | 15.0 | 70.9 | 16.4 | 8.7 | 8.6 | 8.3 | 20.8 | 21.6 | 14.4 | 10.5 | 0.0 | 14.2 | 17.2 | 7.3 | 9.3 | 20.3 | 18.2 | 6.9 | 14.1 | 0.0 | 13.2 | 13.2 |
| (MPI_STRUCT) | 11.0 | 10.6 | 9.8 | 5.1 | 6.1 | 7.2 | 12.0 | 29.1 | 17.2 | 9.6 | 2.7 | 12.8 | 7.7 | 9.5 | 11.5 | 13.4 | 13.5 | 5.0 | 10.2 | 7.9 | 15.6 | 14.3 |
| NECUIUC_CLS-DTCT | 29.7 | 81.8 | 41.9 | 23.1 | 22.4 | 22.0 | 27.8 | 43.2 | 51.8 | 25.9 | 4.5 | 18.5 | 18.0 | 23.5 | 26.9 | 36.6 | 34.8 | 8.8 | 28.3 | 14.0 | 35.5 | 34.7 |
| NECUIUC_SEG | 28.3 | 81.5 | 39.3 | 20.9 | 22.6 | 21.7 | 26.1 | 37.1 | 51.5 | 25.2 | 5.7 | 17.5 | 15.7 | 24.2 | 27.4 | 35.3 | 33.0 | 7.9 | 23.4 | 12.5 | 32.1 | 33.3 |
| (OXFORD_MKL) | 10.9 | 2.0 | 9.2 | 7.7 | 5.3 | 6.1 | 20.1 | 36.7 | 18.2 | 8.5 | 2.9 | 6.5 | 1.4 | 6.2 | 10.9 | 12.2 | 15.5 | 4.2 | 8.9 | 5.1 | 20.8 | 19.7 |
| UC3M_GEN-DIS | 14.5 | 69.8 | 20.8 | 9.7 | 6.3 | 4.3 | 7.9 | 19.7 | 21.8 | 7.7 | 3.8 | 7.5 | 9.6 | 9.5 | 12.3 | 16.5 | 16.4 | 1.5 | 14.2 | 11.0 | 14.1 | 20.3 |
| UCI_LAYEREDSHAPE | 24.7 | 80.7 | 38.3 | 30.9 | 3.4 | 4.4 | 31.7 | 45.5 | 47.3 | 10.4 | 4.8 | 14.3 | 8.8 | 6.1 | 21.5 | 25.0 | 38.9 | 14.8 | 14.4 | 3.0 | 29.1 | 45.5 |
| UCLA_SUPERPIXELCRF | 13.8 | 51.2 | 13.9 | 7.0 | 3.9 | 6.4 | 8.1 | 14.4 | 24.3 | 12.1 | 6.4 | 10.3 | 14.5 | 6.7 | 9.7 | 23.6 | 20.0 | 2.3 | 12.6 | 12.3 | 17.0 | 13.2 |
| (UVA_BAGOFWINDOWS) | 12.6 | 12.3 | 10.9 | 5.9 | 5.4 | 10.7 | 7.8 | 36.4 | 17.6 | 9.9 | 4.6 | 11.7 | 12.9 | 7.2 | 17.8 | 19.1 | 16.3 | 2.0 | 15.5 | 6.7 | 22.2 | 11.0 |
| UoCTTI_LSVM-MDPM | 29.0 | 78.9 | 35.3 | 22.5 | 19.1 | 23.5 | 36.2 | 41.2 | 50.1 | 11.7 | 8.9 | 28.5 | 1.4 | 5.9 | 24.0 | 35.3 | 33.4 | 35.1 | 27.7 | 14.2 | 34.1 | 41.8 |
- Entries in parentheses are synthesized from detection results.
| [mean] | back ground |
aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BERKELEY_POSELETS | 5.5 | 78.5 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 36.3 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| BROOKESMSRC_AHCRF | 24.5 | 79.6 | 40.1 | 9.0 | 17.6 | 1.5 | 20.6 | 34.9 | 29.4 | 24.1 | 6.1 | 13.8 | 28.3 | 13.3 | 9.3 | 31.1 | 23.0 | 17.1 | 18.0 | 24.7 | 36.1 | 37.5 |
| aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ALCALA_AVW | 66.1 | 34.6 | 43.0 | 57.0 | 20.0 | 35.3 | 40.5 | 47.3 | 37.8 | 26.6 | 17.8 | 36.3 | 38.5 | 36.2 | 74.7 | 10.9 | 26.1 | 30.7 | 58.0 | 43.8 |
| ALCALA_LAVW | 65.7 | 31.3 | 44.0 | 53.8 | 16.7 | 39.3 | 41.7 | 43.7 | 35.0 | 20.0 | 19.0 | 31.9 | 40.9 | 37.1 | 73.6 | 19.0 | 23.4 | 22.0 | 56.9 | 45.6 |
| CASIA_SVM-MULTIFEAT | 75.4 | 40.9 | 50.9 | 55.2 | 26.5 | 40.4 | 52.0 | 48.0 | 48.6 | 25.4 | 27.3 | 42.2 | 56.0 | 50.8 | 85.8 | 26.3 | 31.8 | 30.6 | 64.2 | 52.6 |
| CNRS_FUSE-KNN-CTS | 63.6 | 28.3 | 36.9 | 51.7 | 20.2 | 35.8 | 41.2 | 33.8 | 35.8 | 19.8 | 18.6 | 31.4 | 38.3 | 33.6 | 73.6 | 13.5 | 25.5 | 17.5 | 49.4 | 40.3 |
| CNRS_KNN-CT | 54.1 | 18.4 | 31.6 | 45.0 | 16.5 | 23.3 | 38.5 | 31.1 | 26.6 | 18.7 | 14.4 | 26.4 | 23.8 | 28.5 | 71.8 | 14.0 | 10.2 | 15.1 | 36.4 | 34.8 |
| CNRS_KNN-GABOR | 50.4 | 20.3 | 33.3 | 29.8 | 17.9 | 23.9 | 35.9 | 27.3 | 33.7 | 10.6 | 13.1 | 20.8 | 16.3 | 27.9 | 69.8 | 8.2 | 8.7 | 20.2 | 27.1 | 32.8 |
| CNRS_KNN-HRGB | 37.5 | 12.2 | 26.7 | 36.1 | 14.4 | 6.4 | 26.0 | 24.9 | 21.5 | 9.1 | 12.0 | 23.1 | 15.0 | 20.0 | 67.7 | 11.6 | 8.6 | 7.6 | 13.3 | 23.2 |
| CNRS_KNN-OSHL | 60.7 | 23.4 | 26.9 | 44.1 | 16.8 | 35.3 | 31.6 | 27.7 | 31.6 | 14.5 | 15.5 | 22.5 | 33.1 | 22.9 | 67.1 | 16.0 | 23.6 | 19.3 | 38.3 | 35.6 |
| CVC_FLAT | 82.3 | 53.6 | 66.4 | 70.0 | 36.9 | 54.5 | 62.4 | 63.2 | 50.3 | 40.5 | 49.2 | 50.3 | 66.7 | 64.2 | 88.6 | 37.2 | 46.3 | 48.0 | 79.2 | 69.6 |
| CVC_FLAT-HOG-ESS | 83.3 | 57.4 | 67.2 | 68.8 | 39.9 | 55.6 | 66.9 | 63.7 | 50.8 | 34.9 | 47.2 | 47.3 | 67.7 | 66.8 | 88.8 | 40.2 | 46.6 | 49.4 | 79.4 | 71.5 |
| CVC_PLUS | 83.5 | 53.1 | 67.0 | 71.2 | 35.0 | 53.8 | 61.1 | 63.4 | 47.3 | 39.7 | 46.7 | 45.6 | 66.7 | 66.4 | 88.3 | 36.6 | 45.8 | 44.6 | 79.5 | 69.5 |
| FIRSTNIKON_AVGSRKDA | 81.1 | 54.4 | 64.4 | 69.1 | 31.6 | 54.6 | 60.4 | 61.7 | 49.7 | 30.7 | 45.8 | 50.9 | 69.4 | 60.2 | 87.1 | 33.7 | 47.0 | 44.3 | 80.8 | 65.6 |
| FIRSTNIKON_AVGSVM | 81.3 | 53.6 | 63.9 | 68.9 | 32.7 | 51.2 | 59.3 | 61.0 | 50.3 | 34.6 | 47.6 | 50.5 | 67.7 | 60.6 | 86.9 | 32.1 | 46.1 | 44.4 | 78.2 | 66.3 |
| FIRSTNIKON_BOOSTSRKDA | 80.8 | 54.7 | 63.6 | 68.7 | 33.0 | 54.6 | 60.9 | 60.1 | 50.8 | 34.8 | 44.9 | 50.2 | 68.9 | 62.8 | 86.9 | 30.7 | 44.8 | 46.0 | 79.9 | 64.6 |
| FIRSTNIKON_BOOSTSVMS | 81.0 | 53.2 | 63.2 | 68.9 | 32.8 | 51.3 | 60.4 | 60.2 | 51.4 | 36.5 | 47.6 | 50.4 | 67.1 | 61.4 | 86.5 | 31.6 | 42.4 | 46.1 | 77.5 | 64.5 |
| FIRST_L2MKL | 8.0 | 9.1 | 12.8 | 8.0 | 10.3 | 3.3 | 16.3 | 9.6 | 12.9 | 3.4 | 3.4 | 18.7 | 6.0 | 5.7 | 55.4 | 8.6 | 3.0 | 4.7 | 4.6 | 5.9 |
| HAS_FISHSIFT-FISHSEG | 48.0 | 17.9 | 25.7 | 33.4 | 11.9 | 25.0 | 21.7 | 34.9 | 22.0 | 9.3 | 8.2 | 30.0 | 16.3 | 19.1 | 69.1 | 10.5 | 8.9 | 10.7 | 21.2 | 25.5 |
| IIR_SVM-ROI-IC | 70.4 | 41.5 | 45.0 | 51.1 | 22.6 | 36.1 | 48.2 | 46.9 | 41.5 | 23.8 | 29.5 | 35.2 | 54.6 | 46.3 | 78.1 | 18.8 | 34.2 | 37.0 | 63.7 | 54.3 |
| KERLE_SVM-DENSESIFT | 70.8 | 40.3 | 48.4 | 59.3 | 19.8 | 40.3 | 53.1 | 48.1 | 43.5 | 19.8 | 24.2 | 41.8 | 52.3 | 46.3 | 81.8 | 15.9 | 30.5 | 38.2 | 58.0 | 50.3 |
| LEAR_CHI-SVM-MULT | 75.3 | 45.2 | 57.2 | 63.9 | 30.1 | 47.3 | 55.3 | 54.5 | 49.1 | 31.4 | 39.1 | 45.3 | 61.6 | 61.0 | 85.3 | 23.7 | 35.2 | 39.0 | 69.5 | 57.9 |
| LEAR_CHI-SVM-MULT-LOC | 75.6 | 52.2 | 57.4 | 66.0 | 43.2 | 51.9 | 69.2 | 55.8 | 49.8 | 31.8 | 42.2 | 46.7 | 60.6 | 62.8 | 86.4 | 38.0 | 40.4 | 41.0 | 70.1 | 65.4 |
| LEOBEN_DENSESIFT | 72.4 | 41.7 | 57.1 | 61.6 | 28.7 | 48.0 | 54.5 | 52.0 | 45.7 | 28.5 | 24.0 | 44.0 | 55.8 | 53.0 | 82.1 | 19.1 | 29.4 | 43.8 | 64.7 | 55.4 |
| LEOBEN_SCC-200 | 78.2 | 44.2 | 55.9 | 63.1 | 22.6 | 45.7 | 54.9 | 53.2 | 46.6 | 26.7 | 36.5 | 43.5 | 57.3 | 52.8 | 84.8 | 27.6 | 40.9 | 40.3 | 68.6 | 59.9 |
| LEOBEN_SCC-CLS | 77.5 | 44.5 | 57.5 | 63.1 | 30.5 | 42.9 | 57.2 | 54.3 | 46.8 | 30.8 | 36.0 | 44.4 | 56.3 | 56.1 | 85.3 | 29.5 | 44.0 | 40.9 | 67.4 | 59.9 |
| LIG_MIRIM-VPH | 59.8 | 20.1 | 30.4 | 37.1 | 13.7 | 22.3 | 27.7 | 31.3 | 33.1 | 10.6 | 7.0 | 19.3 | 20.5 | 22.5 | 71.2 | 8.8 | 18.6 | 8.2 | 19.3 | 36.5 |
| LIG_MRIM-COLORSIFT | 66.7 | 30.3 | 39.4 | 46.3 | 21.9 | 28.8 | 38.4 | 38.5 | 37.1 | 13.7 | 28.2 | 28.6 | 40.4 | 32.8 | 78.3 | 19.9 | 32.4 | 15.9 | 45.7 | 41.5 |
| LIG_MRIM-FUSION | 68.5 | 32.4 | 42.1 | 51.4 | 24.8 | 35.6 | 42.9 | 41.2 | 42.6 | 18.6 | 28.8 | 32.6 | 48.0 | 40.5 | 79.7 | 19.2 | 34.1 | 18.9 | 56.3 | 44.4 |
| LIP6_HB-SPK-SVM | 74.6 | 42.8 | 49.3 | 59.5 | 27.0 | 43.0 | 53.7 | 50.7 | 45.1 | 24.9 | 33.3 | 43.5 | 61.4 | 50.2 | 82.5 | 14.9 | 34.6 | 34.2 | 66.7 | 58.7 |
| LIP6_SS-SPK-SVM | 77.3 | 46.1 | 54.5 | 63.0 | 27.9 | 48.5 | 55.2 | 56.1 | 46.9 | 26.9 | 41.5 | 45.3 | 63.9 | 54.4 | 84.6 | 19.3 | 38.5 | 37.9 | 74.5 | 60.6 |
| LIRIS_BASELINE | 69.6 | 37.5 | 47.5 | 57.1 | 24.1 | 34.5 | 49.9 | 42.5 | 43.5 | 16.1 | 27.5 | 37.6 | 49.0 | 44.0 | 81.8 | 22.8 | 26.5 | 32.0 | 56.2 | 49.0 |
| LIRIS_EER | 69.8 | 37.6 | 47.0 | 58.8 | 23.6 | 35.3 | 49.3 | 42.8 | 43.8 | 16.1 | 27.6 | 37.8 | 48.3 | 44.0 | 81.7 | 22.5 | 26.1 | 31.9 | 55.6 | 48.3 |
| LIRIS_SOFT-BASELINE | 65.8 | 30.5 | 41.0 | 51.8 | 21.5 | 30.5 | 44.7 | 36.4 | 40.7 | 20.0 | 23.4 | 33.4 | 37.7 | 33.7 | 78.8 | 20.8 | 22.7 | 16.7 | 47.8 | 40.9 |
| LIRIS_SOFT-EER | 66.7 | 30.5 | 41.6 | 52.3 | 21.9 | 31.8 | 44.4 | 36.5 | 40.3 | 20.2 | 23.7 | 33.3 | 37.2 | 33.1 | 79.2 | 20.7 | 22.2 | 16.5 | 46.9 | 40.8 |
| MPI_STRUCT | 73.3 | 42.2 | 46.4 | 52.5 | 21.5 | 49.1 | 52.1 | 50.9 | 32.6 | 29.7 | 40.9 | 42.1 | 60.5 | 60.4 | 74.3 | 22.6 | 34.5 | 44.5 | 64.9 | 50.6 |
| NECUIUC_CDCV | 86.7 | 65.0 | 69.6 | 74.0 | 40.1 | 64.4 | 72.2 | 69.9 | 54.3 | 48.9 | 50.4 | 58.0 | 74.8 | 69.8 | 88.6 | 31.4 | 48.4 | 55.4 | 82.9 | 67.3 |
| NECUIUC_CLS-DTCT | 86.3 | 65.3 | 69.5 | 74.0 | 42.4 | 64.8 | 74.1 | 69.7 | 55.2 | 50.0 | 53.3 | 58.7 | 74.7 | 71.3 | 89.1 | 37.2 | 54.1 | 59.4 | 84.2 | 67.3 |
| NECUIUC_LL-CDCV | 85.4 | 62.8 | 66.5 | 74.0 | 39.7 | 63.4 | 71.0 | 68.3 | 54.8 | 44.7 | 52.0 | 55.6 | 74.0 | 69.7 | 87.9 | 29.7 | 47.0 | 53.2 | 82.9 | 65.9 |
| NECUIUC_LN-CDCV | 86.4 | 64.6 | 69.6 | 73.5 | 38.5 | 63.9 | 71.8 | 69.2 | 53.3 | 47.7 | 50.7 | 58.4 | 73.8 | 68.7 | 88.1 | 30.3 | 47.4 | 56.4 | 83.1 | 67.1 |
| RITSU_AKF | 73.3 | 45.1 | 52.8 | 60.3 | 24.2 | 47.4 | 52.9 | 50.0 | 45.3 | 29.1 | 42.6 | 41.5 | 56.8 | 52.5 | 82.0 | 14.2 | 36.5 | 38.3 | 68.3 | 54.5 |
| RITSU_ASF | 73.3 | 46.3 | 52.2 | 60.2 | 24.0 | 46.1 | 51.6 | 48.6 | 44.9 | 30.1 | 42.0 | 41.8 | 56.8 | 52.0 | 81.4 | 13.5 | 36.5 | 38.6 | 67.2 | 52.6 |
| RITSU_WSF | 74.7 | 46.1 | 51.7 | 60.8 | 27.9 | 48.1 | 51.5 | 49.6 | 44.8 | 30.8 | 43.9 | 43.0 | 58.1 | 52.9 | 82.0 | 14.1 | 35.9 | 38.1 | 67.7 | 54.3 |
| TSINGHUA_ALL-SVM-BOOST | 45.4 | 15.6 | 29.0 | 37.4 | 22.1 | 11.0 | 32.8 | 21.6 | 23.4 | 7.4 | 7.9 | 22.7 | 20.9 | 27.6 | 70.3 | 7.1 | 17.2 | 17.6 | 21.6 | 31.1 |
| TSINGHUA_SVM-SEG-HOG | 30.2 | 5.4 | - | 21.3 | 9.5 | 3.8 | 18.7 | 8.4 | 11.7 | 2.2 | 6.0 | 12.0 | - | - | 64.1 | 11.9 | 2.1 | 5.1 | 6.4 | 8.7 |
| UC3M_GEN-DIS | 67.1 | 39.3 | 34.2 | 39.3 | 24.1 | 37.7 | 49.0 | 38.3 | 41.1 | 22.2 | 25.7 | 33.1 | 39.4 | 46.7 | 78.4 | 18.6 | 13.3 | 19.9 | 45.3 | 41.2 |
| UVASURREY_BASELINE | 81.9 | 53.0 | 64.0 | 67.9 | 33.3 | 53.4 | 61.7 | 60.9 | 53.2 | 40.3 | 51.0 | 51.0 | 68.7 | 64.3 | 87.8 | 33.7 | 45.4 | 45.4 | 77.5 | 66.7 |
| UVASURREY_MKFDA+BOW | 83.4 | 58.9 | 67.4 | 69.4 | 36.8 | 57.9 | 65.9 | 63.2 | 52.9 | 36.0 | 54.0 | 54.1 | 70.5 | 69.3 | 88.9 | 36.3 | 46.5 | 51.2 | 80.5 | 68.7 |
| UVASURREY_TUNECOLORKERNELSEL | 82.6 | 57.7 | 66.8 | 69.2 | 36.3 | 58.4 | 64.8 | 62.1 | 52.7 | 38.0 | 53.3 | 52.9 | 69.9 | 68.7 | 88.5 | 36.3 | 45.4 | 51.5 | 80.4 | 68.6 |
| UVASURREY_TUNECOLORSPECKDA | 83.0 | 58.6 | 67.0 | 69.7 | 37.4 | 57.9 | 65.9 | 61.8 | 52.5 | 37.3 | 54.8 | 53.3 | 69.0 | 69.8 | 88.7 | 36.3 | 46.1 | 50.6 | 80.2 | 66.9 |
| aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|---|
| aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CASIA_SVM-PHOG | 17.8 | 12.6 | 9.8 | 9.9 | - | 18.0 | - | - | 2.5 | - | 3.0 | - | - | - | - | - | - | - | 14.5 | - |
| CASIA_SVM-PHOG+COLOR | 25.3 | 18.0 | 10.4 | 10.8 | 9.5 | 21.6 | 14.1 | 12.4 | 9.8 | 6.9 | 14.0 | 11.0 | 19.1 | 22.9 | 17.1 | 9.3 | 4.1 | 10.1 | 23.2 | 18.6 |
| CVC_HOG-BOW | 31.7 | 28.3 | 10.4 | 10.7 | 17.1 | 21.5 | 22.6 | 15.6 | 14.0 | 13.7 | 19.7 | 13.5 | 26.5 | 30.6 | 20.3 | 10.2 | 18.7 | 15.3 | 25.4 | 31.4 |
| CVC_HOG-BOW-ESS-FLAT | 32.2 | 27.9 | 11.2 | 12.2 | 16.7 | 19.5 | 22.0 | 18.7 | 15.7 | 18.7 | 18.0 | 12.8 | 30.8 | 31.9 | 21.9 | 11.4 | 19.5 | 23.1 | 30.0 | 30.7 |
| LEAR_CHI-SVM-SIFT-HOG | 25.6 | 29.5 | 5.1 | 11.0 | 21.6 | 22.9 | 33.9 | 14.5 | 12.3 | 12.7 | 19.8 | 12.6 | 23.0 | 27.0 | 22.2 | 11.0 | 18.7 | 14.2 | 31.4 | 35.1 |
| LEAR_CHI-SVM-SIFT-HOG-CLS | 26.7 | 30.7 | 11.2 | 12.7 | 22.2 | 24.0 | 34.5 | 15.5 | 13.6 | 15.0 | 20.9 | 11.9 | 27.3 | 27.8 | 22.3 | 12.0 | 24.3 | 16.5 | 31.9 | 37.2 |
| MIZZOU_DEF-HOG-LBP | 8.2 | 20.0 | 10.7 | 13.0 | 28.1 | 26.8 | 36.4 | 25.0 | 18.2 | 11.5 | 12.7 | 13.4 | 39.8 | 42.7 | 42.5 | 12.0 | 24.1 | 9.5 | 1.1 | 34.6 |
| MIZZOU_DEF-HOG-LBP-WOCONTEXT | 19.5 | 20.5 | 7.5 | 11.7 | 27.8 | 25.7 | 36.8 | 21.5 | 17.5 | 16.1 | 18.0 | 16.4 | 39.1 | 41.6 | 42.2 | 11.1 | 23.2 | 14.2 | 32.6 | 35.6 |
| MPI_STRUCT | 34.3 | 14.8 | 10.6 | 14.1 | 5.5 | 23.1 | 13.2 | 28.9 | 9.1 | 13.4 | 20.1 | 20.2 | 25.6 | 31.6 | 12.1 | 9.1 | 10.5 | 24.5 | 38.5 | 10.8 |
| NECUIUC_CLS-DTCT | 42.8 | 27.7 | 12.4 | 11.7 | 10.7 | 36.6 | 31.1 | 30.7 | 8.1 | 19.9 | 29.1 | 23.6 | 39.0 | 39.6 | 24.6 | 7.3 | 25.4 | 24.6 | 30.1 | 29.4 |
| OXFORD_MKL | 41.1 | 38.7 | 17.2 | 18.5 | 23.4 | 31.0 | 35.1 | 29.4 | 17.0 | 21.9 | 27.2 | 16.3 | 37.8 | 44.9 | 24.4 | 11.4 | 23.4 | 24.2 | 43.4 | 41.8 |
| TSINGHUA_SVM-SEG-HOG | 9.1 | - | - | 2.3 | 0.0 | - | 9.1 | - | - | 0.0 | - | 0.4 | - | 9.1 | 1.7 | 0.0 | 0.0 | - | 0.5 | 0.1 |
| TTIWEIZ_NNHOUGH | 21.4 | 20.5 | - | - | - | 8.5 | 23.7 | - | - | 13.8 | - | - | 19.4 | 24.9 | - | - | - | - | - | 26.5 |
| UC3M_GEN-DIS | 21.0 | 11.5 | 10.1 | 9.5 | 0.3 | 15.7 | 11.2 | 20.0 | 1.2 | 9.6 | 12.0 | 15.7 | 16.3 | 20.0 | 12.1 | 9.2 | 9.5 | 11.2 | 12.6 | 5.2 |
| UVA_BAGOFWINDOWS | 25.7 | 16.6 | 10.5 | 9.2 | 3.0 | 20.7 | 14.5 | 32.4 | 1.6 | 11.7 | 17.1 | 23.0 | 28.4 | 29.3 | 13.4 | 4.5 | 9.9 | 25.8 | 29.4 | 9.4 |
| UVA_BOWSEG | 35.1 | - | 6.9 | - | - | 16.0 | - | 32.7 | - | - | 20.3 | - | - | - | - | - | - | 23.7 | 27.9 | - |
| UoCTTI_LSVM-MDPM | 34.9 | 47.1 | 15.4 | 14.4 | 30.1 | 31.8 | 40.9 | 21.0 | 17.9 | 24.7 | 6.7 | 14.1 | 43.1 | 44.3 | 47.6 | 12.8 | 25.0 | 18.0 | 37.5 | 36.7 |
| aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BERKELEY_POSELETS | - | - | - | - | - | - | - | - | - | - | - | - | - | - | 48.7 | - | - | - | - | - |
- Entries in parentheses are synthesized from detection results.
| [mean] | back ground |
aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BONN_SVM-SEGM | 36.2 | 83.5 | 52.8 | 22.4 | 20.7 | 35.8 | 46.1 | 50.5 | 39.9 | 35.3 | 6.1 | 33.1 | 25.2 | 19.7 | 42.7 | 50.6 | 36.8 | 23.5 | 43.6 | 16.7 | 26.8 | 47.8 |
| BROOKESMSRC_AHCRF | 23.2 | 79.2 | 35.7 | 4.3 | 20.6 | 6.3 | 14.2 | 30.6 | 28.8 | 24.1 | 5.6 | 6.6 | 27.0 | 8.7 | 12.9 | 39.6 | 22.1 | 18.3 | 16.9 | 17.3 | 33.9 | 35.3 |
| (CASIA_SVM-PHOG+COLOR) | 10.5 | 24.8 | 6.8 | 4.7 | 6.1 | 6.0 | 16.3 | 20.7 | 7.1 | 7.3 | 3.7 | 5.0 | 5.2 | 4.2 | 11.2 | 18.3 | 12.5 | 8.2 | 14.5 | 5.6 | 14.6 | 18.2 |
| CVC_HOCRF | 34.8 | 79.4 | 56.3 | 26.6 | 40.6 | 36.1 | 27.3 | 48.4 | 37.9 | 23.4 | 9.1 | 21.4 | 10.1 | 24.5 | 41.2 | 56.4 | 32.8 | 26.8 | 39.2 | 21.9 | 41.0 | 31.1 |
| (CVC_HOG-BOW) | 9.3 | 2.5 | 8.0 | 6.4 | 3.5 | 6.0 | 12.1 | 17.5 | 12.5 | 6.5 | 2.1 | 7.3 | 1.4 | 4.3 | 12.4 | 13.6 | 18.4 | 2.7 | 15.4 | 8.0 | 16.7 | 17.3 |
| (CVC_HOG-BOW-ESS-FLAT) | 0.9 | 2.2 | 1.4 | 0.3 | 0.1 | 0.4 | 0.8 | 0.1 | 0.3 | 0.5 | 0.3 | 0.1 | 0.4 | 0.1 | 0.2 | 0.1 | 1.7 | 1.5 | 1.6 | 1.5 | 0.2 | 5.1 |
| (LEAR_CHI-SVM-SIFT-HOG) | 8.2 | 7.2 | 6.6 | 4.4 | 3.8 | 4.8 | 15.1 | 17.2 | 14.6 | 5.3 | 2.4 | 6.5 | 2.7 | 2.8 | 8.1 | 6.1 | 10.2 | 8.3 | 9.8 | 6.9 | 12.6 | 16.8 |
| (LEAR_CHI-SVM-SIFT-HOG-CLS) | 5.4 | 7.0 | 3.1 | 1.9 | 1.1 | 1.1 | 3.5 | 15.1 | 8.7 | 5.2 | 4.3 | 0.0 | 2.8 | 2.3 | 1.7 | 2.9 | 8.8 | 1.1 | 6.3 | 8.1 | 6.6 | 23.1 |
| LEAR_SEGDET | 24.1 | 78.4 | 36.8 | 15.1 | 20.5 | 11.8 | 27.8 | 24.5 | 28.7 | 18.4 | 4.7 | 16.4 | 0.8 | 16.1 | 28.2 | 31.4 | 24.5 | 14.9 | 31.4 | 15.7 | 28.8 | 30.8 |
| (MIZZOU_DEF-HOG-LBP) | 7.8 | 0.9 | 1.0 | 2.7 | 0.0 | 0.0 | 15.7 | 33.7 | 14.3 | 6.5 | 1.1 | 2.0 | 0.1 | 3.8 | 15.7 | 23.2 | 16.6 | 0.8 | 20.8 | 0.0 | 0.0 | 4.9 |
| (MIZZOU_DEF-HOG-LBP-WOCONTEXT) | 5.3 | 0.9 | 1.0 | 3.2 | 0.0 | 0.0 | 21.1 | 6.1 | 11.5 | 2.2 | 1.2 | 4.3 | 0.1 | 2.0 | 10.7 | 0.5 | 13.2 | 1.2 | 15.0 | 2.0 | 3.8 | 12.2 |
| MPI_A2 | 15.0 | 71.8 | 16.1 | 8.7 | 10.5 | 7.6 | 14.6 | 20.8 | 9.3 | 6.0 | 0.0 | 15.4 | 21.1 | 5.6 | 13.2 | 27.7 | 20.4 | 9.6 | 14.3 | 0.0 | 11.0 | 12.5 |
| (MPI_STRUCT) | 11.1 | 10.7 | 7.4 | 5.6 | 5.8 | 9.7 | 9.7 | 27.5 | 11.6 | 9.2 | 2.3 | 13.7 | 9.8 | 9.0 | 12.4 | 15.4 | 15.7 | 6.2 | 13.1 | 6.1 | 16.6 | 14.7 |
| NECUIUC_CLS-DTCT | 28.9 | 81.2 | 34.9 | 25.2 | 22.6 | 22.8 | 29.6 | 37.3 | 40.5 | 26.6 | 5.7 | 18.9 | 17.6 | 22.5 | 28.6 | 36.1 | 37.5 | 11.6 | 31.3 | 8.7 | 32.8 | 35.0 |
| NECUIUC_SEG | 27.5 | 80.9 | 32.4 | 22.6 | 22.8 | 22.5 | 27.4 | 30.7 | 39.5 | 25.5 | 7.1 | 18.7 | 15.6 | 23.1 | 30.2 | 34.8 | 35.2 | 10.2 | 27.7 | 7.3 | 29.9 | 33.2 |
| (OXFORD_MKL) | 11.2 | 2.2 | 8.0 | 8.9 | 4.8 | 6.7 | 22.4 | 35.5 | 17.2 | 9.5 | 2.6 | 6.2 | 1.9 | 5.6 | 12.5 | 13.1 | 15.9 | 4.6 | 9.7 | 4.7 | 21.7 | 21.5 |
| UC3M_GEN-DIS | 13.8 | 69.0 | 13.3 | 9.3 | 6.1 | 5.1 | 6.3 | 15.2 | 18.6 | 8.0 | 6.3 | 6.9 | 8.8 | 7.9 | 10.8 | 17.6 | 18.9 | 2.0 | 17.7 | 9.0 | 12.9 | 21.0 |
| UCI_LAYEREDSHAPE | 23.6 | 80.5 | 36.8 | 29.4 | 4.6 | 3.4 | 33.0 | 33.9 | 41.4 | 8.5 | 6.1 | 16.7 | 7.6 | 4.4 | 19.0 | 26.4 | 41.3 | 8.9 | 18.3 | 4.3 | 27.5 | 44.0 |
| UCLA_SUPERPIXELCRF | 13.1 | 50.8 | 10.6 | 7.0 | 3.7 | 5.6 | 8.9 | 11.3 | 16.4 | 10.7 | 6.0 | 8.1 | 15.7 | 6.3 | 12.2 | 23.7 | 21.8 | 2.4 | 13.5 | 7.8 | 18.8 | 13.8 |
| (UVA_BAGOFWINDOWS) | 12.9 | 12.4 | 8.7 | 6.1 | 6.0 | 12.3 | 7.6 | 39.8 | 12.8 | 10.7 | 4.8 | 10.9 | 15.0 | 8.3 | 17.1 | 18.4 | 18.0 | 2.2 | 19.5 | 6.3 | 23.1 | 11.2 |
| UoCTTI_LSVM-MDPM | 29.2 | 78.7 | 34.8 | 23.4 | 20.0 | 29.0 | 37.3 | 27.8 | 42.3 | 11.4 | 9.0 | 31.9 | 0.5 | 5.9 | 27.2 | 38.1 | 36.7 | 37.1 | 33.6 | 16.5 | 32.1 | 40.5 |
- Entries in parentheses are synthesized from detection results.
| [mean] | back ground |
aero plane |
bicycle | bird | boat | bottle | bus | car | cat | chair | cow | dining table |
dog | horse | motor bike |
person | potted plant |
sheep | sofa | train | tv/ monitor |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BERKELEY_POSELETS | 5.5 | 78.1 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 37.8 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| BROOKESMSRC_AHCRF | 23.3 | 79.2 | 32.4 | 8.2 | 19.9 | 1.2 | 18.5 | 32.9 | 23.0 | 24.4 | 7.8 | 4.0 | 31.8 | 10.0 | 9.7 | 33.6 | 24.4 | 21.7 | 17.8 | 19.0 | 34.2 | 36.5 |
| Abbreviation | Method | Affiliation | Contributors | Descriptiorn |
|---|---|---|---|---|
| ALCALA_AVW | AVW | University of Alcalá - GRAM | Roberto J. López Sastre, Saturnino Maldonado Bascón | Aggregating Visual Words (AVW) obtained from different runs of the K-means algorithm over SIFT descriptors. We use a clustering aggregation approach to build more efficient codebooks for category-level object recognition. An SVM with histogram intersection kernel is used for classification. Please, see attached README for more details ;-) |
| ALCALA_LAVW | LAVW | University of Alcalá - GRAM | Roberto J. López Sastre, Saturnino Maldonado Bascón | With this submission we present an approach to perform LOCAL Aggregation of Visual Words (LAVW). Our aim is to incorporate into a traditional clustering aggregation approach [1] a factor to lead the algorithm to merge only those visual words that are near in the image domain and not only in the descriptor space. Please see attached README. |
| BERKELEY_POSELETS | Poselets | U.C. Berkeley / Adobe | Lubomir Bourdev, Subhransu Maji, Jitendra Malik | The method is described in the paper: Poselets: Body Part Detectors Trained Using 3D Human Pose Annotations available here: http://www.eecs.berkeley.edu/~lbourdev/poselets |
| BONN_SVM-SEGM | SVM-Segm | University of Bonn | João Carreira Fuxin Li Cristian Sminchisescu | We compute multiple segmentations of each image, and learn to classify them into one of the 20 classes, using only the annotated 1500 segmentation images. We employ a SVM framework with multiple kernels, that encode shape and appearance with different levels of invariance. To segment a test image, we use simple heuristics to decide which segments to keep, and which label they should get. |
| BROOKESMSRC_AHCRF | AH-CRF | Oxford Brookes University, Microsoft Research Cambridge | Lubor Ladicky Chris Russell Pushmeet Kohli Philip H.S. Torr | Hierarchical CRF with pixel, superpixel, segment and label preference layer. Pixel layer contains unary based on shape filters of 3 features (SIFT / Opponent SIFT / Texton), intensity based pairwise potential, superpixel contains consistency enforcing unary potential and pairwise potential based on EMD distance of histograms of colour, segment layer contains potential based on histograms of features. Solved with alpha expansion. |
| CASIA_SVM-MULTIFEAT | SVM-MultiFeature | Institute of Automation, Chinese Academy of Sciences | Gang Cheng | Svm classifier with multi features. Trained using 5-fold cross-validation. Re-trained on full train+val set with fixed parameters. |
| CASIA_SVM-PHOG | SVM-phog | Institute of Automation, Chinese Academy of Sciences. | Rongguo Zhang, Baihua Xiao, Linbo Zhang, Chunheng Wang. | Lib-SVM classifier with phog features. Trained using 5-fold cross-validation. Re-trained on full train+val set with fixed parameters. Got the candidate detection windows by segmentation method instead of the sliding way. |
| CASIA_SVM-PHOG+COLOR | SVM-phog+color | Institute of Automation, Chinese Academy of Sciences. | Rongguo Zhang, Baihua Xiao, linbo Zhang, Chunheng Wang. | Lib-SVM classifier with phog and colorphog features. Trained using 5-fold cross-validation. Re-trained on full train+val set with fixed parameters. |
| CNRS_FUSE-KNN-CTS | FUSE_KNN_CTS | Laborory of Informatics of Grenoble - CNRS | Georges Quénot | Late fusion of KNN classifiers outputs from Opponent SIFT filtered by Harris- Laplace and from early fusion of color (4x4x4 RGB histogram) and texture (8 orientations x 5 scales Gabor transform). Trained on trainval and evaluated on test. 6-fold validation is used of KNN parameter tuning. |
| CNRS_KNN-CT | KNN_CT | Laborory of Informatics of Grenoble - CNRS | Georges Quénot | KNN classifier output from early fusion of color (4x4x4 RGB histogram) and texture (8 orientations x 5 scales Gabor transform). Trained on trainval and evaluated on test. 6-fold validation is used of KNN parameter tuning. |
| CNRS_KNN-GABOR | KNN_GABOR | Laborory of Informatics of Grenoble - CNRS | Georges Quénot | KNN classifier output from texture (8 orientations x 5 scales Gabor transform). Trained on trainval and evaluated on test. 6-fold validation is used of KNN parameter tuning. |
| CNRS_KNN-HRGB | KNN_HRGB | Laborory of Informatics of Grenoble - CNRS | Georges Quénot | KNN classifier output from color (4x4x4 RGB histogram). Trained on trainval and evaluated on test. 6-fold validation is used of KNN parameter tuning. |
| CNRS_KNN-OSHL | KNN_OSHL | Laborory of Informatics of Grenoble - CNRS | Georges Quénot | KNN classifier output from Opponent SIFT filtered by Harris- Laplace. Trained on trainval and evaluated on test. 6-fold validation is used of KNN parameter tuning. |
| CVC_FLAT | CVC-Flat | Computer Vision Center Barcelona | Fahad Shahbaz Khan, Joost van de Weijer, Andrew Bagdanov, Noha Elfiky, David Rojas, Marco Pedersoli, xavier Boix, Pep Gonfaus, Hany salahEldeen, Robert Benavente, Jordi Gonzalez, and Maria Vanrell. | Features Detection : Harris Laplace, Color Boosted HarisLaplace, Dense Multi-scale Grid, Blob, and Color Boosted Blob detector. Features Extraction: SIFT , Hue, Color names, Opp-SIFT, C-SIFT and RGSIFT, GIST and spatial pyramids. Vocabulary Construtction: agglomerative information bottleneck method [Vedaldi]. Color Attention: For combining color and shape, we use our top-down color attention method [Khan ICCV 2009]. Learning: We use intersection kernel for learning. |
| CVC_FLAT-HOG-ESS | CVC-flat+detection | Computer Vision Center Barcelona | Fahad Shahbaz Khan, Joost van de Weijer, Andrew Bagdanov, Noha Elfiky, David Rojas, Marco Pedersoli, xavier Boix, Pep Gonfaus, Hany salahEldeen, Robert Benavente, Jordi Gonzalez, and Maria Vanrell. | We combine the CVC_flat image classification results, with the scores of the HOG pyramids detector, and ESS detector. For combination we use a method closely related to [Harzallah ICCV 2009]. |
| CVC_HOCRF | CVC | Computer Vision Center - Barcelona | Xavier Boix, Josep Maria Gonfaus, Fahad Kahn, Joost van de Weijer, Andrew Bagdanov, Marco Pedersoli, Jordi González, Joan Serrat | -High Order CRF. -Learning of the probabilistic model. -Superpixel based. -BOW on superpixels + BOW on context. -Global classifier used in competition 1. -Superpixel classifiers are trained only on his usual context of the object. -Features: SIFT + RGB histogram. |
| CVC_HOG-BOW | CVC-det | Computer Vision Center Barcelona | Marco Pedersoli, Andrew Bagdanov, Joost van de Weijer, Fahad Shahbaz Khan, Davide Modolo, Jordi Gonzàlez and Juan José Villanueva. | Base Detector: Dense sliding window search using HOG pyramids. Iterative linear SVM re-training with re-localization of the positives examples and hard negatives selection. Refinement: Re-scoring of the first 100 detection using SVM intersection kernel (trained adding random negative examples) and additional features (BOW pyramid). |
| CVC_HOG-BOW-ESS-FLAT | CVC-det-fusion | Computer Vision Center Barcelona | Marco Pedersoli, Joost van de Weijer, Andrew Bagdanov, Fahad Shahbaz Khan, Davide Modolo, Jordi Gonzàlez and Juan José Villanueva. | Base Detector: CVC-det and ESS detectors are combined using an OR strategy Combination: detector output is combined with classification CVC-flat according to [Harzallah ICCV09] |
| CVC_PLUS | CVC-plus | Computer Vision Center Barcelona | Fahad Shahbaz Khan, Joost van de Weijer, Andrew Bagdanov, Noha Elfiky, David Rojas, Marco Pedersoli, xavier Boix, Pep Gonfaus, Hany salahEldeen, Robert Benavente, Jordi Gonzalez, and Maria Vanrell. | All of CVC-flat + additional SIFT pyramids histograms. Classification has been performed with SVM with Chi-square Kernel. |
| FIRSTNIKON_AVGSRKDA | bindernakajima_avgsrkda | Fraunhofer FIRST, Nikon Corp. | Alexander Binder, Fraunhofer FIRST Shinichi Nakajima, Nikon Corp. Motoaki Kawanabe, Fraunhofer FIRST | same features as 177 but with average kernel srkda and kernel width opt |
| FIRSTNIKON_AVGSVM | bindernakajima_avgsvm | Fraunhofer FIRST, Nikon Corp. | Alexander Binder, Fraunhofer FIRST Shinichi Nakajima, Nikon Corp. Motoaki Kawanabe, Fraunhofer FIRST | same features as 177 but with average kernel svm and kernel width opt |
| FIRSTNIKON_BOOSTSRKDA | bindernakajima_boostsrkda | Fraunhofer FIRST, Nikon Corp. | Alexander Binder, Fraunhofer FIRST Shinichi Nakajima, Nikon Corp. Motoaki Kawanabe, Fraunhofer FIRST | same features as 177 but with boosting over srkda and kernel width opt |
| FIRSTNIKON_BOOSTSVMS | bindernakajima_boostsvms | Fraunhofer FIRST, Nikon Corp. | Alexander Binder, Fraunhofer FIRST Shinichi Nakajima, Nikon Corp. Motoaki Kawanabe, Fraunhofer FIRST | same as 177 features, boosted svms |
| FIRST_L2MKL | bindernakajima_L2MKL | Fraunhofer FIRST | Alexander Binder, Fraunhofer FIRST Shinichi Nakajima, Nikon Corp. Motoaki Kawanabe, Fraunhofer FIRST | same features, but L2MKL |
| HAS_FISHSIFT-FISHSEG | FishSIFT_cluSeg | Data Mining and Web Search Group, Hungarian Academy of Sciences | Balint Daroczy, Andras Benczur, Dora Erdos, Zsolt Fekete, Istvan Petras | Logistic regression with Fisher kernel SIFT and Fisher kernel Segmentation. Trained on train+val. |
| IIR_SVM-ROI-IC | SVM-Combined ROI and image codebooks | Institute for Infocomm Research, Singapore | GAO Yan, LIM Joo Hwee, LI Yiqun, GOH Hanlin | feature: HSVsift + spatial pyramid; classifer: SVM with histogram intersection kernel. |
| KERLE_SVM-DENSESIFT | SVM-SIFT | Marc Kerle | Libsvm classifier with dense SIFT features. Parameter selection using the val set. | |
| LEAR_CHI-SVM-MULT | CHI-SVM + Multiple features | INRIA Grenoble | Adrien Gaidon Cordelia Schmid Marcin Marszalek | Same as 2008's Adrien Gaidon Submission. |
| LEAR_CHI-SVM-MULT-LOC | CHI-SVM + Multiple features + Comb. with localiz. | INRIA Grenoble | Hedi Harzallah Cordelia Schmid Frederic Jurie Adrien Gaidon | see classification method of "Combining efficient object localization and image classification, ICCV'09" |
| LEAR_CHI-SVM-SIFT-HOG | CHI-SVM-SIFT-HOG | INRIA Grenoble | Hedi Harzallah Cordelia Schmid and Frederic Jurie | 2 stage classification (linear SVM, then chi square SVM) with HOG and dense SIFT features. |
| LEAR_CHI-SVM-SIFT-HOG-CLS | CHI-SVM-SIFT-HOG + Comb. with classification | INRIA Grenoble | Hedi Harzallah Cordelia Schmid Frederic Jurie Adrien Gaidon | see localization method of "Combining efficient object localization and image classification, ICCV'09" |
| LEAR_SEGDET | Combo-Seg-Det | LEAR, INRIA Grenoble | Tingting Jiang, Cordelia Schmid, Frederic Jurie, Hedi Harzallah, Adrien Gaidon | SVM classifier based on Bag of Words on masks. Trained with Intersection Kernel SVM and improved by detection |
| LEOBEN_DENSESIFT | MUL-baseline | University of Leoben | Martin Antenreiter, Thomas Jaksch, Peter Auer | Svm classifier with dense SIFT features. Trained using 5-fold cross-validation. Re-trained on full train+val set with fixed parameters. |
| LEOBEN_SCC-200 | SCC-200 | University of Leoben | Martin Antenreiter, Thomas Jaksch, Peter Auer | SVM classifiers with various SIFT features. Trained using 5-fold cross-validation. Re-trained on full train set with fixed parameters. Combined classifier trained using 5-fold cross-validation on val set. Re-trained on full val set with fixed parameters. |
| LEOBEN_SCC-CLS | SCC-CLS | University of Leoben | Martin Antenreiter, Thomas Jaksch, Peter Auer | SVM classifiers with various SIFT features. Trained using 5-fold cross-validation. Re-trained on full train set with fixed parameters. Class-wise selection of classifiers for combination. Several combined classifiers trained using 5-fold cross-validation on val set, re-trained on full val set with fixed parameters. Choice of final classifier per class based on CV performance. |
| LIG_MIRIM-VPH | LIG_MRIM_VPH | LIG, University Joseph Fourier - Grenoble 1 | Rami Albatal , Philippe Mulhem, Yves Chiaramella | We use Harris-Laplace interest regions detector with rgSIFT descriptor, then we construct sets of interest regions (Visual Phrases) in each image in BOVW representation, then we use SVM classifier for classifing individual sets. |
| LIG_MRIM-COLORSIFT | LIG_MRIM_ColorSift | LIG, University Joseph Fourier - Grenoble 1 | Rami Albatal , Philippe Mulhem, Yves Chiaramella | Harris-Laplace detector + rgSift descriptor, codebook of 4000 bin created using k-means clustering (classical method proposed by Koen E. A. van de Sande) |
| LIG_MRIM-FUSION | LIG_MRIM_Fusion | LIG, University Joseph Fourier - Grenoble 1 | Rami Albatal , Philippe Mulhem, Yves Chiaramella, Georges Quénot | We fuse 5 Svm classifier : 1- rgSift (codebook 4000 bin), 2- Opponent Sift (codebook 4000 bin), 3- hg104 (early fusion of rbg histogram and gabor histogram), 4- Visual Phrase classifier (used in LIG_MRIM_VPH method), 5- another Visual phrase mehode based on a filtering of interest regions. Fusion based on linear combination of SVM results. |
| LIP6_HB-SPK-SVM | HB_SPK_SVM | UPMC LIP6 | David Picard, Nicolas Thome, Matthieu Cord | SVM classifier over SPK of dense SIFT based visual codebook (4000 entries). |
| LIP6_SS-SPK-SVM | SS_SPK_SVM | UPMC LIP6 | David Picard, Nicolas Thome, Matthieu Cord | SVM classifier with SPK of dense SIFT using 4000 words and semi soft assignment. |
| LIRIS_BASELINE | LIRIS_Baseline | LIRIS, Ecole Centrale de Lyon, France | Chao ZHU, Huanzhang FU, Charles-Edmond BICHOT, Emmanuel Dellandrea, Liming CHEN | LibSVM classifier with dense SIFT, Color (CCV, CH, CM), Texture (LBP, GLCM) and Shape (EH) features. A vocabulary of 4000 codewords is created for bag-of-features model of SIFT, hard assignment is adapted. The chi square distance is uesd as the kernel of SVM for all kinds of features. Scores of different kinds of features are fused as the final classification score by mean. Trained on 'train' set, and parameters are tuned on 'val' set. |
| LIRIS_EER | LIRIS_EER | LIRIS, Ecole Centrale de Lyon, France | Chao ZHU, Huanzhang FU, Charles-Edmond BICHOT, Emmanuel Dellandrea, Liming CHEN | LibSVM classifier with dense SIFT, Color (CCV, CH, CM), Texture (LBP, GLCM) and Shape (EH) features. A vocabulary of 4000 codewords is created for bag-of-features model of SIFT, hard assignment is adapted. The chi square distance is uesd as the kernel of SVM for all kinds of features. Scores of different kinds of features are fused as the final classification score according to their EER (Equal Error Rate). Trained on 'train' set, and parameters are tuned on 'val' set. |
| LIRIS_SOFT-BASELINE | LIRIS_soft_Baseline | LIRIS, Ecole Centrale de Lyon, France | Chao ZHU, Huanzhang FU, Charles-Edmond BICHOT, Emmanuel Dellandrea, Liming CHEN | LibSVM classifier with dense SIFT, Color (CCV, CH, CM), Texture (LBP, GLCM) and Shape (EH) features. A vocabulary of 4000 codewords is created for bag-of-features model of SIFT, soft assignment is adapted. The chi square distance is uesd as the kernel of SVM for all kinds of features. Scores of different kinds of features are fused as the final classification score by mean. Trained on 'train' set, and parameters are tuned on 'val' set. |
| LIRIS_SOFT-EER | LIRIS_soft_EER | LIRIS, Ecole Centrale de Lyon, France | Chao ZHU, Huanzhang FU, Charles-Edmond BICHOT, Emmanuel Dellandrea, Liming CHEN | LibSVM classifier with dense SIFT, Color (CCV, CH, CM), Texture (LBP, GLCM) and Shape (EH) features. A vocabulary of 4000 codewords is created for bag-of-features model of SIFT, soft assignment is adapted. The chi square distance is uesd as the kernel of SVM for all kinds of features. Scores of different kinds of features are fused as the final classification score according to their EER (Equal Error Rate). Trained on 'train' set, and parameters are tuned on 'val' set. |
| MIZZOU_DEF-HOG-LBP | DEF-HOG-LBP | The University of Missouri | Xiaoyu Wang Wei Gong Xutao Lv Tony Han | Deformable model with HOG-LBP features. |
| MIZZOU_DEF-HOG-LBP-WOCONTEXT | DEF-HOG-LBP-WOCONTEXT | The University of Missouri | Xiaoyu Wang, Wei Gong, Xutao Lv, Tony X.Han | Deformable models using HOG-LBP features. Bounding box are trained for prediction. |
| MPI_A2 | MPI-A2 | Max Planck Institute for Biological Cybernetics | Sebastian Nowozin, Christoph Lampert | Log-linear tree-structured CRF on superpixel representation with simple features and many parameters. All parameters jointly trained using conditional log-likelihood. Training exclusively using seg-trainval (1499 images). Model selection performed using seg-train/seg-val, parameters fixed and one final training on seg-trainval. |
| MPI_STRUCT | MPIstruct | Max Planck Institute for Biological Cybernetics | Christoph Lampert | Structured Regression with linear kernel for bounding box prediction. Nonlinear SVM trained on trainval predictions for re-ranking. |
| NECUIUC_CDCV | NEC_UIUC_CLS_CDCV | NEC Laboratories America and University of Illinois at Urbana-Champaign | Yihong Gong, Fengjun Lv, Jinjun Wang, Chen Wu, Wei Xu, Jianchao Yang, Kai Yu, Xi Zhou, Thomas Huang | Nonlinear classifiers implemented via linear classification methods on nonlinear mapping of SIFT features, obtained by unsupervised local linear & nonlinear coding methods. Class dependent cross validation. |
| NECUIUC_CLS-DTCT | NEC_UIUC_CLS&DTCT | NEC Laboratories America and University of Illinois at Urbana-Champaign | Yihong Gong, Fengjun Lv, Jinjun Wang, Chen Wu, Wei Xu, Jianchao Yang, Kai Yu, Xi Zhou, Thomas Huang | (1) Classification: nonlinear classifiers implemented via linear classification methods on nonlinear mapping of SIFT features, obtained by unsupervised local linear & nonlinear coding methods. The result is further enhanced by an multi-instance learner using results from the detection task; (2) Detection: Probabilistic inference on segments of an image to find candidate class-specific foregrounds. Linear classifiers used to scan those candidates to get the details of bounding boxes. |
| NECUIUC_LL-CDCV | NEC_UIUC_CLS_LL_CDCV | NEC Laboratories America and University of Illinois at Urbana-Champaign | Yihong Gong, Fengjun Lv, Jinjun Wang, Chen Wu, Wei Xu, Jianchao Yang, Kai Yu, Xi Zhou, Thomas Huang | Nonlinear classifiers implemented via linear classification methods on nonlinear mapping of SIFT features, obtained by unsupervised local linear coding methods. |
| NECUIUC_LN-CDCV | NEC_UIUC_CLS_LN_CDCV | NEC Laboratories America and University of Illinois at Urbana-Champaign | Yihong Gong, Fengjun Lv, Jinjun Wang, Chen Wu, Wei Xu, Jianchao Yang, Kai Yu, Xi Zhou, Thomas Huang | Nonlinear classifiers implemented via linear classification methods on nonlinear mapping of SIFT features, obtained by unsupervised local nonlinear coding methods. |
| NECUIUC_SEG | NEC_UIUC_SEG | NEC Laboratories America and University of Illinois at Urbana-Champaign | Yihong Gong, Fengjun Lv, Jinjun Wang, Chen Wu, Wei Xu, Jianchao Yang, Kai Yu, Xi Zhou, Thomas Huang | Based on our detection method. |
| OXFORD_MKL | VGG-MKL | Oxford University | A. Vedaldi, V. Gulshan, M. Varma, A. Zisserman | The method is described in "Multiple Kernels for Object Detection" A. Vedaldi, V. Gulshan, M. Varma, and A. Zisserman ICCV'09 |
| RITSU_AKF | RitsOM_AKF | Ritsumeikan University | Xian-Hua Han, Yen-Wei Chen, Xiang Ruan | LibSVM classifier with Bag-of-feature and PHOG shape feature. Fusion the multiple descriptprs with avarage Kernel. Trained on full train+val set with fixed parameters. |
| RITSU_ASF | RitsOM_ASF | Ritsumeikan University | Xian-Hua Han, Yen-Wei Chen, Xiang Ruan | LibSVM classifier with Bag-of-feature and PHOG shape feature. Fusion the multiple descriptprs with avarage similarites. Trained on full train+val set with fixed parameters. |
| RITSU_WSF | RitsOM_WSF | Ritsumeikan University | Xian-Hua Han, Yen-Wei Chen, Xiang Ruan | LibSVM classifier with Bag-of-feature and PHOG shape feature. Fusion the multiple descriptprs with Weighted similarites. Trained on full train+val set with fixed parameters. |
| TSINGHUA_ALL-SVM-BOOST | all_fused | IMG, Tsinghua University | Binbin Cao | 34 features are used. Models are trained by rankboost. Results are fused by sum. |
| TSINGHUA_SVM-SEG-HOG | SVM-SEG-HOG | National Tsing Hua University | Bo-Cheng Chen Kevin Soong Terry Luo Chia-Mao Hung Shin-Hung Ji | Trained on full train+val set with fixed parameters. |
| TTIWEIZ_NNHOUGH | NN-Hough | Toyota Technological Institute at Chicago and Weizmann Institute of Science | Gregory Shakhnarovich, Daniel Glasner | We find nearest neighbors of densely sampled patches (on multiscale pyramid) in a database of patches inside bounding boxes of object in training data. Each match votes for a bounding box, with a weight learned for all DB patches by gradient descent. Scores based on these votes are used to select a small set of candidate detections. Then posterior is estimated using logistic regression and scale prior. Finally, detections for different views and with different descriptors are pooled. |
| UC3M_GEN-DIS | UC3M_Gen_Dis | Universidad Carlos III de Madrid | Iván González Díaz | This is a two-layered approach. The bottom layer is implemented using a generative model, (an extension of the one in attached document), that incorporates the spatial structure of visual documents. Its purpose to work at the region level and generate a set of probabilities for both documents and regions (detection/segmentation). The upper layer is implemented by discriminative models (SVMs) that classify images based on the output probabilities of the generative algorithm (classification). |
| UCI_LAYEREDSHAPE | Layered shape models | University of California at Irvine | Charless Fowlkes Sam Hallman Deva Ramanan Yi Yang | We use the detector from Felzenswalb et al to train models using the 2009 det trainval data. We then train our segmentation system using the detector and the 2009 seg trainval data. Our segmentation system is based on category-specific part-based shape priors. We integrate these priors with instance-specific color models estimated for each putative bounding box. Models from high-scoring bounding boxes are composited together, and are used to label individual superpixels with a category label. |
| UCLA_SUPERPIXELCRF | superpixelcrf | UCLA | Brian Fulkerson, Andrea Vedaldi, Stefano Soatto | This is a direct implementation of Class Segmentation and Object Localization with Superpixel Neighborhoods, B. Fulkerson, A. Vedaldi, and S. Soatto. ICCV 2009. It was trained on train+val, with the number of neighbors fixed at 3. Source code will be available during the conference. See http://vision.ucla.edu/~brian/superpixelcrf.html for more details. |
| UVASURREY_BASELINE | UvASurrey-Base | University of Amsterdam and University of Surrey | Koen van de Sande, Fei Yan, Atif Tahir, Jasper Uijlings, Mark Barnard, Hongping Cai, Theo Gevers, Arnold Smeulders, Krystian Mikolajczyk, Josef Kittler | This run is directly comparable to the SoftColorSIFT run from last year. It uses the same dense sampling, Harris-Laplace keypoints and 4 Color SIFT descriptors, equal kernel weights and Support Vector Machines. It uses color descriptor software available from http://www.colordescriptors.com |
| UVASURREY_MKFDA+BOW | MK-FDA + BoW | University of Amsterdam, University of Surrey | K. Sande J. Uijlings M. Barnard F. Yan H. Cai P. Koniusz A. Tahir K. Mikolajczyk J. Kittler T. Gevers A. Smeulders | learning the weights of kernels that are generated using various sampling techniques and descriptors. the kernels weights are learnt in non-sparse multiple kernel fisher discriminant analysis. |
| UVASURREY_TUNECOLORKERNELSEL | UvASurrey-TunableColorKernelSelection | University of Amsterdam and University of Surrey | Koen van de Sande, Fei Yan, Atif Tahir, Jasper Uijlings, Mark Barnard, Hongping Cai, Theo Gevers, Arnold Smeulders, Krystian Mikolajczyk, Josef Kittler | This run uses new tunable color filters on top of the existing color descriptors, several SIFT variations and a combination of kernel selection and multiple kernel fisher discriminant analysis. |
| UVASURREY_TUNECOLORSPECKDA | UvASurrey-TunableColorSpectralKDA | University of Amsterdam and University of Surrey | Koen van de Sande, Atif Tahir, Fei Yan, Jasper Uijlings, Mark Barnard, Hongping Cai, Theo Gevers, Arnold Smeulders, Krystian Mikolajczyk, Josef Kittler | This run uses new tunable color filters on top of the existing color descriptors, several SIFT variations and a special form of spectral kernel discriminant analysis. |
| UVA_BAGOFWINDOWS | UvA-BagOfWindows | University of Amsterdam | Koen E.A. van de Sande Jasper R.R. Uijlings Theo Gevers Arnold W.M. Smeulders | Efficient subwindow search with dense SIFT, a bag-of-windows classifier and weighting using the UvASurrey-Base run. Descriptor software available from http://www.colordescriptors.com |
| UVA_BOWSEG | UvABoWSeg | University of Amsterdam | J.R.R. Uijlings K. van de Sande A.W.M. Smeulders R.J.H. Scha | The order of the images is created using our BoW classifier. Afterwards, a segmentation algorithm proposes a limited set of bounding boxes which is evaluated using again BoW with Spatial Pyramid and chi^2 SVM kernel |
| UoCTTI_LSVM-MDPM | LSVM-MDPM | University of Chicago and TTI-C | Pedro Felzenszwalb (UChicago), Ross Girshick (UChicago), David McAllester (TTI-C) | Our submission is based on [1]. Each class is represented by a mixture of deformable part models (6 components with 6 parts per class). We also have a binary mask associated to each component of each class to generate pixel-level segmentations from detections. The models were trained from bounding boxes. The segmentation masks were trained from segmentations. [1] Felzenszwalb, Girshick, McAllester, Ramanan, "Object Detection with Discriminatively Trained Part Based Models", PAMI (preprint) |
| (CL_13oct09) | CL_13oct09 | - | - | MPIhybrid |