Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Omni-Scale Feature Learning for Person Re-Identification

About

As an instance-level recognition problem, person re-identification (ReID) relies on discriminative features, which not only capture different spatial scales but also encapsulate an arbitrary combination of multiple scales. We call features of both homogeneous and heterogeneous scales omni-scale features. In this paper, a novel deep ReID CNN is designed, termed Omni-Scale Network (OSNet), for omni-scale feature learning. This is achieved by designing a residual block composed of multiple convolutional streams, each detecting features at a certain scale. Importantly, a novel unified aggregation gate is introduced to dynamically fuse multi-scale features with input-dependent channel-wise weights. To efficiently learn spatial-channel correlations and avoid overfitting, the building block uses pointwise and depthwise convolutions. By stacking such block layer-by-layer, our OSNet is extremely lightweight and can be trained from scratch on existing ReID benchmarks. Despite its small model size, OSNet achieves state-of-the-art performance on six person ReID datasets, outperforming most large-sized models, often by a clear margin. Code and models are available at: \url{https://github.com/KaiyangZhou/deep-person-reid}.

Kaiyang Zhou, Yongxin Yang, Andrea Cavallaro, Tao Xiang• 2019

Related benchmarks

TaskDatasetResultRank
Person Re-IdentificationMarket1501 (test)
Rank-1 Accuracy94.8
1264
Person Re-IdentificationDuke MTMC-reID (test)
Rank-188.6
1018
Person Re-IdentificationMarket 1501
mAP84.9
999
Person Re-IdentificationDukeMTMC-reID
Rank-1 Acc88.6
648
Image ClassificationCIFAR-100--
622
Person Re-IdentificationMSMT17 (test)
Rank-1 Acc78.9
499
Image ClassificationImageNet ILSVRC-2012 (val)
Top-1 Accuracy75.5
405
Person Re-IdentificationMSMT17
mAP0.529
404
Person Re-IdentificationMarket-1501 (test)
Rank-194.8
384
Person Re-IdentificationCUHK03 (Detected)
Rank-1 Accuracy72.3
219
Showing 10 of 45 rows

Other info

Code

Follow for update