Deep learning

Electrical engineering

Physics

Channel (broadcasting)

Pattern recognition (psychology)

Organic chemistry

Overfitting

Feature (linguistics)

Block (permutation group theory)

Computer network

Mathematics

Discriminative model

Fuse (electrical)

Machine learning

Engineering

Philosophy

Artificial intelligence

Set (abstract data type)

Chemistry

Programming language

Geometry

Code (set theory)

Scale (ratio)

Layer (electronics)

Computer science

Linguistics

Quantum mechanics

Artificial neural network

Feature learning

Convolutional neural network

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 callse 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 feature 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 both pointwise and depthwise convolutions. By stacking such blocks layer-by-layer, our OSNet is extremely lightweight and can be trained from scratch on existing ReID benchmarks. Despite its small model size, our OSNet achieves state-of-the-art performance on six person-ReID datasets. Code and models are available at: https://github.com/KaiyangZhou/deep-person-reid.

Omni-Scale Feature Learning for Person Re-Identification