In Defense of the Triplet Loss for Person Re-Identification
In the past few years, the field of computer vision has gone through a revolution fueled mainly by the advent of large datasets and the adoption of deep convolutional neural networks for end-to-end learning. The person re-identification subfield is no exception to this. Unfortunately, a prevailing belief in the community seems to be that the triplet loss is inferior to using surrogate losses (classification, verification) followed by a separate metric learning step. We show that, for models trained from scratch as well as pretrained ones, using a variant of the triplet loss to perform end-to-end deep metric learning outperforms most other published methods by a large margin.
PDF AbstractCode
Datasets
Market-1501
CUHK03
DukeMTMC-reID
MARS
Results from the Paper
Ranked #3 on
Person Re-Identification
on CUHK03
(Rank-5 metric)
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
|---|---|---|---|---|---|---|
| Person Re-Identification | CUHK03 | TriNet | Rank-1 | 89.63 | # 5 | |
| Rank-5 | 99.01 | # 3 | ||||
| Person Re-Identification | DukeMTMC-reID | TriNet | Rank-1 | 72.44 | # 75 | |
| mAP | 53.50 | # 78 | ||||
| Person Re-Identification | Market-1501 | LuNet | Rank-1 | 81.38 | # 111 | |
| Rank-5 | 92.34 | # 15 | ||||
| mAP | 60.71 | # 120 | ||||
| Person Re-Identification | Market-1501 | TriNet | Rank-1 | 84.92 | # 105 | |
| Rank-5 | 94.21 | # 13 | ||||
| mAP | 69.14 | # 110 | ||||
| Person Re-Identification | Market-1501 | LuNet (RK) | Rank-1 | 84.59 | # 106 | |
| Rank-5 | 91.89 | # 16 | ||||
| mAP | 75.62 | # 103 | ||||
| Person Re-Identification | Market-1501 | TriNet (RK) | Rank-1 | 86.67 | # 101 | |
| Rank-5 | 93.38 | # 14 | ||||
| mAP | 81.07 | # 98 | ||||
| Person Re-Identification | MARS | TriNet (RK) | mAP | 77.43 | # 14 | |
| Rank-1 | 81.21 | # 13 | ||||
| Rank-5 | 90.76 | # 8 | ||||
| Person Re-Identification | MARS | LuNet (RK) | mAP | 73.68 | # 15 | |
| Rank-1 | 78.48 | # 15 | ||||
| Rank-5 | 88.74 | # 10 | ||||
| Person Re-Identification | MARS | LuNet | mAP | 60.48 | # 19 | |
| Rank-1 | 75.56 | # 17 | ||||
| Rank-5 | 89.70 | # 9 | ||||
| Person Re-Identification | MARS | TriNet | mAP | 67.70 | # 18 | |
| Rank-1 | 79.80 | # 14 | ||||
| Rank-5 | 91.36 | # 7 |
