Joint Detection and Identification Feature Learning for Person Search

TL;DR

This paper introduces a unified deep learning framework for person search that jointly detects and identifies individuals in whole scene images, utilizing a novel loss function and a large-scale benchmark dataset.

Contribution

It proposes a joint detection and identification model with an Online Instance Matching loss and provides a new large-scale dataset for person search evaluation.

Findings

The joint framework outperforms separate detection and re-identification methods.

The OIM loss converges faster and yields better accuracy than Softmax loss.

The new dataset enables more comprehensive evaluation of person search methods.

Abstract

Existing person re-identification benchmarks and methods mainly focus on matching cropped pedestrian images between queries and candidates. However, it is different from real-world scenarios where the annotations of pedestrian bounding boxes are unavailable and the target person needs to be searched from a gallery of whole scene images. To close the gap, we propose a new deep learning framework for person search. Instead of breaking it down into two separate tasks---pedestrian detection and person re-identification, we jointly handle both aspects in a single convolutional neural network. An Online Instance Matching (OIM) loss function is proposed to train the network effectively, which is scalable to datasets with numerous identities. To validate our approach, we collect and annotate a large-scale benchmark dataset for person search. It contains 18,184 images, 8,432 identities, and 96,143 pedestrian bounding boxes. Experiments show that our framework outperforms other separate approaches, and the proposed OIM loss function converges much faster and better than the conventional Softmax loss.