Diverse Part Discovery: Occluded Person Re-identification with Part-Aware Transformer

Feng Wu; Jianfeng He; Tianzhu Zhang; Xiang Liu; Yongdong Zhang; Yulin Li

RECORD

Diverse Part Discovery: Occluded Person Re-identification with Part-Aware Transformer

Yulin Li, Jianfeng He, Tianzhu Zhang, Xiang Liu, Yongdong Zhang, Feng Wu

Computer Vision and Pattern Recognition · Submitted 8 Jun 2021

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim Extensive experimental results on six challenging benchmarks for three tasks (occluded, partial and holistic Re-ID) demonstrate that our proposed PAT performs favorably against stat-of-the-art methods.

how it is stated · asserted 1.00 read the record →

Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID2106.04095 (version 1)

Pith ID3U3LR2ZJ, Pith's permanent record of this paper and its review

CitationFeng Wu, Jianfeng He, Tianzhu Zhang, Xiang Liu, Yongdong Zhang, Yulin Li. "Diverse Part Discovery: Occluded Person Re-identification with Part-Aware Transformer." arXiv:2106.04095 (2021). https://pith.science/paper/3U3LR2ZJ

Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 5,629 input tokens.

What the abstract itself states, as probabilities: claims first 0.95 · states limitation 0.14 · human participants 0.15 · controlled comparison 0.46 · randomized assignment 0.01 · states quantitative result 0.31 · compares with prior results 0.92 · mentions code or data availability 0.02.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

Occluded person re-identification (Re-ID) is a challenging task as persons are frequently occluded by various obstacles or other persons, especially in the crowd scenario. To address these issues, we propose a novel end-to-end Part-Aware Transformer (PAT) for occluded person Re-ID through diverse part discovery via a transformer encoderdecoder architecture, including a pixel context based transformer encoder and a part prototype based transformer decoder. The proposed PAT model enjoys several merits. First, to the best of our knowledge, this is the first work to exploit the transformer encoder-decoder architecture for occluded person Re-ID in a unified deep model. Second, to learn part prototypes well with only identity labels, we design two effective mechanisms including part diversity and part discriminability. Consequently, we can achieve diverse part discovery for occluded person Re-ID in a weakly supervised manner. Extensive experimental results on six challenging benchmarks for three tasks (occluded, partial and holistic Re-ID) demonstrate that our proposed PAT performs favorably against stat-of-the-art methods.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	Occluded person re-identification (Re-ID) is a challenging task as persons are frequently occluded by various obstacles or other persons, especially in the crowd scenario.	background 1.00	0.05	0.02	asserted 1.00
s1	To address these issues, we propose a novel end-to-end Part-Aware Transformer (PAT) for occluded person Re-ID through diverse part discovery via a transformer encoderdecoder architecture, including a pixel context based transformer encoder and a part prototype based transformer decoder.	method 0.51	0.92	0.11	asserted 0.99
s2	The proposed PAT model enjoys several merits.	contribution 0.95	0.31	0.02	asserted 0.91
s3	First, to the best of our knowledge, this is the first work to exploit the transformer encoder-decoder architecture for occluded person Re-ID in a unified deep model.	contribution 0.97	0.70	0.06	asserted 0.99
s4	Second, to learn part prototypes well with only identity labels, we design two effective mechanisms including part diversity and part discriminability.	method 0.64	0.93	0.04	asserted 1.00
s5	Consequently, we can achieve diverse part discovery for occluded person Re-ID in a weakly supervised manner.	contribution 0.58	0.88	0.16	asserted 0.55
s6	Extensive experimental results on six challenging benchmarks for three tasks (occluded, partial and holistic Re-ID) demonstrate that our proposed PAT performs favorably against stat-of-the-art methods.	result 1.00	0.96	0.79	asserted 1.00

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.98.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper

Original abstract

Occluded person re-identification (Re-ID) is a challenging task as persons are frequently occluded by various obstacles or other persons, especially in the crowd scenario. To address these issues, we propose a novel end-to-end Part-Aware Transformer (PAT) for occluded person Re-ID through diverse part discovery via a transformer encoderdecoder architecture, including a pixel context based transformer encoder and a part prototype based transformer decoder. The proposed PAT model enjoys several merits. First, to the best of our knowledge, this is the first work to exploit the transformer encoder-decoder architecture for occluded person Re-ID in a unified deep model. Second, to learn part prototypes well with only identity labels, we design two effective mechanisms including part diversity and part discriminability. Consequently, we can achieve diverse part discovery for occluded person Re-ID in a weakly supervised manner. Extensive experimental results on six challenging benchmarks for three tasks (occluded, partial and holistic Re-ID) demonstrate that our proposed PAT performs favorably against stat-of-the-art methods.

Human review

How Pith checks papers → · Full audit of this paper →

How the literature uses it

What this paper does with the arXiv papers it cites

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

What it cites · 1 sentence without a confident label

Uses the method or tool of 2 papers

For classifiers, as in [ref] we use a batch normalization layer [ref] and a fully connected layer followed by a softmax function.

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift · Implementation Details · 1.00
The residual connections followed by the layer normalization [ref] are also applied.

Layer Normalization · Pixel Context based Transformer Encoder · 0.98

Cites as background 4 papers

(3) The attention based methods [ref] exploit attention mechanisms to localize discriminative human parts.

A Novel Teacher-Student Learning Framework For Occluded Person Re-Identification · Introduction · 0.88
(1) The hand-crafted splitting based methods divide the image or feature map into small patches [ref] or rigid stripes [ref] and then extract part features from the local patches or stripes.

Recognizing Partial Biometric Patterns · Introduction · 0.83
Recently, person Re-ID has drawn a growing amount of interest from academia and industry, and various methods have been proposed [ref] .

In Defense of the Triplet Loss for Person Re-Identification · Introduction · 0.94

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.