OH-Former: Omni-Relational High-Order Transformer for Person Re-Identification
This work addresses the challenge of person re-identification for surveillance and security applications, representing an incremental improvement over existing transformer methods.
The paper tackled the problem of person re-identification by proposing OH-Former, a transformer-based model that improves performance by modeling high-order feature relations and enhancing local details, achieving state-of-the-art results on datasets like Market-1501 and DukeMTMC.
Transformers have shown preferable performance on many vision tasks. However, for the task of person re-identification (ReID), vanilla transformers leave the rich contexts on high-order feature relations under-exploited and deteriorate local feature details, which are insufficient due to the dramatic variations of pedestrians. In this work, we propose an Omni-Relational High-Order Transformer (OH-Former) to model omni-relational features for ReID. First, to strengthen the capacity of visual representation, instead of obtaining the attention matrix based on pairs of queries and isolated keys at each spatial location, we take a step further to model high-order statistics information for the non-local mechanism. We share the attention weights in the corresponding layer of each order with a prior mixing mechanism to reduce the computation cost. Then, a convolution-based local relation perception module is proposed to extract the local relations and 2D position information. The experimental results of our model are superior promising, which show state-of-the-art performance on Market-1501, DukeMTMC, MSMT17 and Occluded-Duke datasets.