Learning Context Graph for Person Search

Yan, Yichao; Zhang, Qiang; Ni, Bingbing; Zhang, Wendong; Xu, Minghao; Yang, Xiaokang

doi:10.1109/cvpr.2019.00226

Cited by 178 publications

(119 citation statements)

References 54 publications

Supporting

Mentioning

119

Contrasting

Order By: Relevance

“…State-of-the Art (SOTA): Joint Detection and Re-ID Models. To the best of our knowledge, there is only a few work on the joint training of detector and re-identifier for person search task, such as the OIM model [51], the end-to-end model (initialized model) [57], NPSM [49], IAN [67], RCAA [68] and Context Graph [50]. Therefore, these end-to-end person search methods are selected as the SOTA competitor of our I-Net and DC-I-Net.…”

Section: Compared Methodsmentioning

confidence: 99%

“…For the latter, NPSM [49] introduced a LSTM based endto-end person search method which automatically reduces the region containing the target person from a given image. Yan et al [50] firstly introduced the GCN in person search for exploring the relation between instances in an image based on the context information and achieved SOTA performance. Xiao et al [51] jointly trained the detection and person re-identification parts during the training phase, in which a classical OIM loss function is introduced for person re-identification.…”

Section: Person Searchmentioning

confidence: 99%

See 1 more Smart Citation

Tasks Integrated Networks: Joint Detection and Retrieval for Image Search

Zhang

Yang

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

The traditional object (person) retrieval (re-identification) task aims to learn a discriminative feature representation with intra-similarity and inter-dissimilarity, which supposes that the objects in an image are manually or automatically pre-cropped exactly. However, in many real-world searching scenarios (e.g., video surveillance), the objects (e.g., persons, vehicles, etc.) are seldom accurately detected or annotated. Therefore, object-level retrieval becomes intractable without bounding-box annotation, which leads to a new but challenging topic, i.e. image-level search with multi-task integration of joint detection and retrieval. In this paper, to address the image search issue, we first introduce an end-to-end Integrated Net (I-Net), which has three merits: 1) A Siamese architecture and an on-line pairing strategy for similar and dissimilar objects in the given images are designed. Benefited by the Siamese structure, I-Net learns the shared feature representation, because, on which, both object detection and classification tasks are handled. 2) A novel on-line pairing (OLP) loss is introduced with a dynamic feature dictionary, which alleviates the multi-task training stagnation problem, by automatically generating a number of negative pairs to restrict the positives. 3) A hard example priority (HEP) based softmax loss is proposed to improve the robustness of classification task by selecting hard categories. The shared feature representation of I-Net may restrict the task-specific flexibility and learning capability between detection and retrieval tasks. Therefore, with the philosophy of divide and conquer, we further propose an improved I-Net, called DC-I-Net, which makes two new contributions: 1) two modules are tailored to handle different tasks separately in the integrated framework, such that the task specification is guaranteed. 2) A class-center guided HEP loss (C 2 HEP) by exploiting the stored class centers is proposed, such that the intra-similarity and inter-dissimilarity can be captured for ultimate retrieval. Extensive experiments on famous image-level search oriented benchmark datasets, such as CUHK-SYSU dataset and PRW dataset for person search and the large-scale WebTattoo dataset for tattoo search, demonstrate that the proposed DC-I-Net outperforms the state-of-the-art tasks-integrated and tasks-separated image search models.

show abstract

Section: Compared Methodsmentioning

confidence: 99%

Section: Person Searchmentioning

confidence: 99%

Tasks Integrated Networks: Joint Detection and Retrieval for Image Search

Zhang

Yang

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

show abstract

“…Besides the context-based post-processing person re-id methods mentioned above, some context-based learning methods are proposed in recent years, in which the contextual information is utilized based on graph theory with an end-toend deep learning framework [7], [16], [17], [8]. By contrast, the proposed method has a good compatibility and high efficiency, and can utilize various content-based methods as baseline and enhance the performance of person re-id by the contextual information in the unsupervised setting with low computation complexity.…”

Section: B Context-based Person Re-id Methodsmentioning

confidence: 99%

“…Several context-based person re-id methods have been proposed recently [7], [8], [9], [10], [4], [11], in which the contextual information of samples is introduced to assist the computation of the pairwise similarity by the ranking list comparison [12], [13], [14], [15], or by building an end-to-end deep learning framework based on graph theory [16], [17]. In these methods, the samples in one batch or dataset are utilized as the context sample to assist person re-id, and the set-to-set comparison or graph theory based deep learning are made for person re-id.…”

Section: Introductionmentioning

confidence: 99%

Progressive Bilateral-Context Driven Model for Post-Processing Person Re-Identification

Cao

Chen

Dou

et al. 2021

IEEE Trans. Multimedia

View full text Add to dashboard Cite

Most existing person re-identification methods compute pairwise similarity by extracting robust visual features and learning the discriminative metric. Owing to visual ambiguities, these content-based methods that determine the pairwise relationship only based on the similarity between them, inevitably produce a suboptimal ranking list. Instead, the pairwise similarity can be estimated more accurately along the geodesic path of the underlying data manifold by exploring the rich contextual information of the sample. In this paper, we propose a lightweight post-processing person re-identification method in which the pairwise measure is determined by the relationship between the sample and the counterpart's context in an unsupervised way. We translate the point-to-point comparison into the bilateral point-to-set comparison. The sample's context is composed of its neighbor samples with two different definition ways: the first order context and the second order context, which are used to compute the pairwise similarity in sequence, resulting in a progressive post-processing model. The experiments on four large-scale person re-identification benchmark datasets indicate that (1) the proposed method can consistently achieve higher accuracies by serving as a post-processing procedure after the content-based person re-identification methods, showing its stateof-the-art results, (2) the proposed lightweight method only needs about 6 milliseconds for optimizing the ranking results of one sample, showing its high-efficiency.

show abstract

“…Graph models are utilized in several computer vision tasks, and Graph Neural Networks (GNN) is introduced in [35] to model the relations between graph nodes, and a large number of the variants [36]- [41] are proposed. In recent, Re-ID methods [42]- [46] combined with graph models are also explored. Cheng et al [42] formulate the structured distance relationships into the graph Laplacian form to take advantages of the relationships among training samples.…”

Section: Related Workmentioning

confidence: 99%

Few-Shot Deep Adversarial Learning for Video-Based Person Re-Identification

Wang

Yin

et al. 2020

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Recent years have witnessed a great development of deep learning based video person re-identification (Re-ID).A key factor for video person Re-ID is how to effectively construct discriminative video feature representations for the robustness to many complicated situations like occlusions. Recent part-based approaches employ spatial and temporal attention to extract the representative local features. While the correlations between the parts are ignored in the previous methods, to leverage the relations of different parts, we propose an innovative adaptive graph representation learning scheme for video person Re-ID, which enables the contextual interactions between the relevant regional features. Specifically, we exploit pose alignment connection and feature affinity connection to construct an adaptive structure-aware adjacency graph, which models the intrinsic relations between graph nodes. We perform feature propagation on the adjacency graph to refine the original regional features iteratively, the neighbor nodes information is taken into account for part feature representation. To learn the compact and discriminative representations, we further propose a novel temporal resolution-aware regularization, which enforces the consistency among different temporal resolutions for the same identities. We conduct extensive evaluations on four benchmarks, i.e. iLIDS-VID, PRID2011, MARS, and DukeMTMC-VideoReID, the experimental results achieve the competitive performance which demonstrates the effectiveness of our proposed method.

show abstract

Learning Context Graph for Person Search

Cited by 178 publications

References 54 publications

Tasks Integrated Networks: Joint Detection and Retrieval for Image Search

Tasks Integrated Networks: Joint Detection and Retrieval for Image Search

Progressive Bilateral-Context Driven Model for Post-Processing Person Re-Identification

Few-Shot Deep Adversarial Learning for Video-Based Person Re-Identification

Contact Info

Product

Resources

About