A formal approach to good practices in Pseudo-Labeling for Unsupervised Domain Adaptive Re-Identification

Fabian Dubourvieux (1, 2) , Romaric Audigier (1) , Angélique Loesch (1) , Samia Ainouz (2) , Stéphane Canu (2)

1 DIASI (CEA, LIST) - Département Intelligence Ambiante et Systèmes Interactifs
2 LITIS - Laboratoire d'Informatique, du Traitement de l'Information et des Systèmes

The use of pseudo-labels prevails to tackle Unsupervised Domain Adaptive (UDA) Re-Identification (re-ID) with the best performance. Indeed, this family of approaches has given rise to several UDA re-identification-specific frameworks, which are effective. In these works, research directions to improve Pseudo-Labeling UDA re-ID performance are varied and primarily based on intuition and experiments: refining pseudo-labels, reducing the impact of errors in pseudo-labels... It can be hard to deduce from them general good practices that we can implement in any Pseudo-Labeling method to improve its performance consistently. We propose a new theoretical view on Pseudo-Labeling UDA re-ID to address this fundamental question. The contributions are threefold: (i) A novel theoretical framework for Pseudo-Labeling UDA re-ID, formalized through a new general learning upper bound on the UDA re-ID performance. (ii) General good practices for Pseudo-Labeling are directly deduced from the interpretation of the proposed theoretical framework to improve the target re-ID performance. (iii) Extensive experiments on challenging person and vehicle cross-dataset re-ID tasks, showing consistent performance improvements for various state-of-the-art methods and various proposed implementations of good practices.

Domaines: Informatique [cs]

Fabian Dubourvieux, Romaric Audigier, Angélique Loesch, Samia Ainouz, Stéphane Canu. A formal approach to good practices in Pseudo-Labeling for Unsupervised Domain Adaptive Re-Identification. Computer Vision and Image Understanding, 2022, 223, pp.103527. ⟨10.1016/j.cviu.2022.103527⟩. ⟨hal-04125235⟩