OH-Former: Omni-Relational High-Order Transformer for Person Re-Identification

Chen, Xianing; Xu, Jialang; Xu, Jiale; Gao, Shenghua

doi:10.48550/arxiv.2109.11159

Cited by 4 publications

(4 citation statements)

References 47 publications

(60 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many works (Zhu et al 2021;Chen et al 2021;Lai, Chai, and Wei 2021) devote to extract partial region representation by transformer. For example, AAformer (Zhu et al 2021) uses the additional learnable vectors of 'part tokens' to learn the part representations by clustering the patch embeddings into several groups and integrates the part into the self-attention for alignment.…”

Section: Representation Learning In Re-idmentioning

confidence: 99%

DC-Former: Diverse and Compact Transformer for Person Re-Identification

Li¹,

Zhang²,

Wang³

et al. 2023

Preprint

View full text Add to dashboard Cite

In person re-identification (re-ID) task, it is still challenging to learn discriminative representation by deep learning, due to limited data. Generally speaking, the model will get better performance when increasing the amount of data. The addition of similar classes strengthens the ability of the classifier to identify similar identities, thereby improving the discrimination of representation. In this paper, we propose a Diverse and Compact Transformer (DC-Former) that can achieve a similar effect by splitting embedding space into multiple diverse and compact subspaces. Compact embedding subspace helps model learn more robust and discriminative embedding to identify similar classes. And the fusion of these diverse embeddings containing more fine-grained information can further improve the effect of re-ID. Specifically, multiple class tokens are used in vision transformer to represent multiple embedding spaces. Then, a self-diverse constraint (SDC) is applied to these spaces to push them away from each other, which makes each embedding space diverse and compact. Further, a dynamic weight controller (DWC) is further designed for balancing the relative importance among them during training. The experimental results of our method are promising, which surpass previous state-of-the-art methods on several commonly used person re-ID benchmarks. https://github.com/ant-research/Diverseand-Compact-Transformer

show abstract

Section: Representation Learning In Re-idmentioning

confidence: 99%

DC-Former: Diverse and Compact Transformer for Person Re-Identification

Li¹,

Zhang²,

Wang³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Following previous works [9,37] in person ReID, we use the common cross-entropy (CE) loss and triplet loss to train the model. The CE loss has no label smoothing operation and can be defined as follow:…”

Section: The Objective Of Trainingmentioning

confidence: 99%

An adaptive cross‐scale transformer based on graph signal processing for person re‐identification

Zhou

Hou

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

Extracting robust feature representation is one of the key challenges for person reidentification (ReID) task. Although convolution neural network (CNN)-based methods have achieved great success, they still cannot handle the part occlusion and misalignment caused by limited receptive field. Recently, pure transformer models have shown its power in the person ReID task. However, current transformer models adopt patches of equal-scale as input, and cannot solve the problem of cross-scale interaction properly. To overcome this problem, an adaptive cross-scale transformer from a perspective of the graph signal, named ACSFormer, is proposed. Specifically, the self-attention module is first treated as an undirected fully connected graph. And then, "node variation" is introduced as an indicator to adaptively merge neighbourhood tokens. To the best of the authors' knowledge, their ACSFormer is the first work to attempt to combine pure transformers and graph signal processing in the field of person ReID. Extensive evaluations are conducted on three person ReID datasets to validate the performance of ACSFormer. Experiments demonstrate that this ACSFormer performs on par with state-of-the-art CNN-based methods and consistently improves transformer-based baseline, for example, surpassing ViT-baseline by 2.5%, 2.7% and 4.8% mAP on Market1501, DukeMTMC-reID and MSMT17, respectively.

show abstract

“…Since they lack inductive bias, they indeed learn inductive biases from amounts of data implicitly and lag behind CNNs in the low data regime [15]. Recently, some works try to introduce CNNs into vision transformers explicitly [9,11,18,30,[51][52][53]. However, their forcefully modified structure destroyed the intrinsic properties in transformers.…”

Section: Related Workmentioning

confidence: 99%

DearKD: Data-Efficient Early Knowledge Distillation for Vision Transformers

Chen¹,

Cao²,

Zhong³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Transformers are successfully applied to computer vision due to their powerful modeling capacity with selfattention. However, the excellent performance of transformers heavily depends on enormous training images. Thus, a data-efficient transformer solution is urgently needed. In this work, we propose an early knowledge distillation framework, which is termed as DearKD, to improve the data efficiency required by transformers. Our DearKD is a two-stage framework that first distills the inductive biases from the early intermediate layers of a CNN and then gives the transformer full play by training without distillation. Further, our DearKD can be readily applied to the extreme data-free case where no real images are available. In this case, we propose a boundary-preserving intradivergence loss based on DeepInversion to further close the performance gap against the full-data counterpart. Extensive experiments on ImageNet, partial ImageNet, data-free setting and other downstream tasks prove the superiority of DearKD over its baselines and state-of-the-art methods.

show abstract

OH-Former: Omni-Relational High-Order Transformer for Person Re-Identification

Cited by 4 publications

References 47 publications

DC-Former: Diverse and Compact Transformer for Person Re-Identification

DC-Former: Diverse and Compact Transformer for Person Re-Identification

An adaptive cross‐scale transformer based on graph signal processing for person re‐identification

DearKD: Data-Efficient Early Knowledge Distillation for Vision Transformers

Contact Info

Product

Resources

About