arXiv:2201.09604

End-to-end Person Search Sequentially Trained on Aggregated Dataset

Angelique Loesch, Jaonary Rabarisoa, Romaric Audigier

제출 2022-01-24

원문 초록

In video surveillance applications, person search is a challenging task consisting in detecting people and extracting features from their silhouette for re-identification (re-ID) purpose. We propose a new end-to-end model that jointly computes detection and feature extraction steps through a single deep Convolutional Neural Network architecture. Sharing feature maps between the two tasks for jointly describing people commonalities and specificities allows faster runtime, which is valuable in real-world applications. In addition to reaching state-of-the-art accuracy, this multi-task model can be sequentially trained task-by-task, which results in a broader acceptance of input dataset types. Indeed, we show that aggregating more pedestrian detection datasets without costly identity annotations makes the shared feature maps more generic, and improves re-ID precision. Moreover, these boosted shared feature maps result in re-ID features more robust to a cross-dataset scenario.

함께 살펴볼 논문

공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.

arXiv:2207.11898공통 태그 Pedestrian Detection, Person Search공식 코드 · Papers with Code

Domain Adaptive Person Search

arXiv:2103.11617공통 태그 Pedestrian Detection, Person Search공식 코드 · Papers with Code

Anchor-Free Person Search

토론

이 논문과 연결된 포럼 글이 아직 없습니다.