Journal
COMPUTER VISION - ECCV 2022, PT XIV
Volume 13674, Issue -, Pages 336-353Publisher
SPRINGER INTERNATIONAL PUBLISHING AG
DOI: 10.1007/978-3-031-19781-9_20
Keywords
Unstructured feature decoupling network; Vehicle reid; Transformer-based decoupling head; Cluster-based decoupling constraint
Funding
- National Science Foundation of China [NSFC 61906194]
- National Key R&D Program of China [2021YFF0602101]
- Alibaba Group through Alibaba Research Intern Program
Ask authors/readers for more resources
This paper proposes an Unstructured Feature Decoupling Network (UFDN) to address the problem of misalignment of features caused by pose and viewpoint variances in Vehicle Re-Identification (ReID). UFDN aligns features without the need for additional annotation and introduces a cluster-based decoupling constraint to learn diverse but aligned decoupled features. Experimental results demonstrate that UFDN achieves state-of-the-art performance on popular Vehicle ReID benchmarks with both CNN and Transformer backbones.
The misalignment of features caused by pose and viewpoint variances is a crucial problem in Vehicle Re-Identification (ReID). Previous methods align the features by structuring the vehicles from predefined vehicle parts (such as logos, windows, etc.) or attributes, which are inefficient because of additional manual annotation. To align the features without requirements of additional annotation, this paper proposes a Unstructured Feature Decoupling Network (UFDN), which consists of a transformer-based feature decomposing head (TDH) and a novel cluster-based decoupling constraint (CDC). Different from the structured knowledge used in previous decoupling methods, we aim to achieve more flexible unstructured decoupled features with diverse discriminative information as shown in Fig. 1. The self-attention mechanism in the decomposing head helps the model preliminarily learn the discriminative decomposed features in a global scope. To further learn diverse but aligned decoupled features, we introduce a cluster-based decoupling constraint consisting of a diversity constraint and an alignment constraint. Furthermore, we improve the alignment constraint into a modulated one to eliminate the negative impact of the outlier features that cannot align the clusters in semantics. Extensive experiments show the proposed UFDN achieves state-of-the-art performance on three popular Vehicle ReID benchmarks with both CNN and Transformer backbones.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available