Aggregating Deep Pyramidal Representations for Person Re-Idenfitication
Learning discriminative, view-invariant and multi-scale representations of person appearance with different se- mantic levels is of paramount importance for person Re- Identification (Re-ID). A surge of effort has been spent by the community to learn deep Re-ID models capturing a holistic single semantic level feature representation. To improve the achieved results, additional visual attributes and body part-driven models have been considered. How- ever, these require extensive human annotation labor or de- mand additional computational efforts. We argue that a pyramid-inspired method capturing multi-scale information may overcome such requirements. Precisely, multi-scale stripes that represent visual information of a person can be used by a novel architecture factorizing them into latent discriminative factors at multiple semantic levels. A multi- task loss is combined with a curriculum learning strategy to learn a discriminative and invariant person representation which is exploited for triplet-similarity learning. Results on three benchmark Re-ID datasets demonstrate that better performance than existing methods are achieved (e.g., more than 90% accuracy on the Duke-MTMC dataset).
PDFTasks
Datasets
Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
---|---|---|---|---|---|---|
Person Re-Identification | DukeMTMC-reID | PyrNet (+ReRank) | Rank-1 | 90.3 | # 28 | |
mAP | 87.7 | # 18 | ||||
Person Re-Identification | DukeMTMC-reID | PyrNet | Rank-1 | 87.1 | # 49 | |
mAP | 74.0 | # 56 | ||||
Person Re-Identification | Market-1501 | PyrNet (multi-shot+ReRank) | Rank-1 | 96.1 | # 21 | |
mAP | 94.0 | # 18 | ||||
Person Re-Identification | Market-1501 | PyrNet (single-shot) | Rank-1 | 93.6 | # 72 | |
mAP | 81.7 | # 83 | ||||
Person Re-Identification | Market-1501 | PyrNet (multi-shot) | Rank-1 | 95.2 | # 54 | |
mAP | 86.7 | # 69 | ||||
Person Re-Identification | Market-1501 | PyrNet (single-shot+ReRank) | Rank-1 | 94.6 | # 65 | |
mAP | 91.4 | # 27 |