SA-Person: Text-Based Person Retrieval with Scene-aware Re-ranking

• URL: http://arxiv.org/abs/2505.24466v2
• Date: Thu, 26 Jun 2025 12:46:10 GMT
• Title: SA-Person: Text-Based Person Retrieval with Scene-aware Re-ranking
• Authors: Yingjia Xu, Jinlin Wu, Zhen Chen, Daming Gao, Yang Yang, Zhen Lei, Min Cao,
• Abstract summary: Existing methods primarily emphasize appearance-based cross-modal retrieval, often neglecting the contextual information embedded within the scene.<n>We introduce SCENEPERSON-13W, a large-scale dataset featuring over 100,000 scenes with rich annotations covering both pedestrian appearance and environmental cues.<n>In the first stage, it performs discriminative appearance grounding by aligning textual cues with pedestrian-specific regions.<n>In the second stage, it introduces SceneRanker, a training-free, scene-aware re-ranking method leveraging multimodal large language models to jointly reason over pedestrian appearance and the global scene context.
• Score: 20.515788520147453
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Text-based person retrieval aims to identify a target individual from a gallery of images based on a natural language description. It presents a significant challenge due to the complexity of real-world scenes and the ambiguity of appearance-related descriptions. Existing methods primarily emphasize appearance-based cross-modal retrieval, often neglecting the contextual information embedded within the scene, which can offer valuable complementary insights for retrieval. To address this, we introduce SCENEPERSON-13W, a large-scale dataset featuring over 100,000 scenes with rich annotations covering both pedestrian appearance and environmental cues. Based on this, we propose SA-Person, a two-stage retrieval framework. In the first stage, it performs discriminative appearance grounding by aligning textual cues with pedestrian-specific regions. In the second stage, it introduces SceneRanker, a training-free, scene-aware re-ranking method leveraging multimodal large language models to jointly reason over pedestrian appearance and the global scene context. Experiments on SCENEPERSON-13W validate the effectiveness of our framework in challenging scene-level retrieval scenarios. The code and dataset will be made publicly available.

Related papers

• Enhancing Visual Representation for Text-based Person Searching [9.601697802095119]
  VFE-TPS is a Visual Feature Enhanced Text-based Person Search model.<n>It introduces a pre-trained backbone CLIP to learn basic multimodal features.<n>It constructs Text Guided Masked Image Modeling task to enhance the model's ability of learning local visual details.
  arXiv  Detail & Related papers  (2024-12-30T01:38:14Z)
• In Defense of Lazy Visual Grounding for Open-Vocabulary Semantic Segmentation [50.79940712523551]
  We present lazy visual grounding, a two-stage approach of unsupervised object mask discovery followed by object grounding. Our model requires no additional training yet shows great performance on five public datasets.
  arXiv  Detail & Related papers  (2024-08-09T09:28:35Z)
• Text2Place: Affordance-aware Text Guided Human Placement [26.041917073228483]
  This work tackles the problem of realistic human insertion in a given background scene termed as textbfSemantic Human Placement. For learning semantic masks, we leverage rich object-scene priors learned from the text-to-image generative models. The proposed method can generate highly realistic scene compositions while preserving the background and subject identity.
  arXiv  Detail & Related papers  (2024-07-22T08:00:06Z)
• StableSemantics: A Synthetic Language-Vision Dataset of Semantic Representations in Naturalistic Images [5.529078451095096]
  understanding the semantics of visual scenes is a fundamental challenge in Computer Vision. Recent advancements in text-to-image frameworks have led to models that implicitly capture natural scene statistics. Our work presents StableSemantics, a dataset comprising 224 thousand human-curated prompts, processed natural language captions, over 2 million synthetic images, and 10 million attention maps corresponding to individual noun chunks.
  arXiv  Detail & Related papers  (2024-06-19T17:59:40Z)
• VLLMs Provide Better Context for Emotion Understanding Through Common Sense Reasoning [66.23296689828152]
  We leverage the capabilities of Vision-and-Large-Language Models to enhance in-context emotion classification. In the first stage, we propose prompting VLLMs to generate descriptions in natural language of the subject's apparent emotion. In the second stage, the descriptions are used as contextual information and, along with the image input, are used to train a transformer-based architecture.
  arXiv  Detail & Related papers  (2024-04-10T15:09:15Z)
• BEV-TSR: Text-Scene Retrieval in BEV Space for Autonomous Driving [46.84729450920804]
  We propose the BEV-TSR framework which leverages descriptive text as an input to retrieve corresponding scenes in the Bird's Eye View space. We employ a large language model (LLM) to extract the semantic features of the text inputs and incorporate knowledge graph embeddings to enhance the semantic richness of the language embedding. Experimental results on the multi-level nuScenes-Retrieval show that BEV-TSR achieves state-of-the-art performance, e.g., 85.78% and 87.66% top-1 accuracy on scene-to-text and text-to-scene
  arXiv  Detail & Related papers  (2024-01-02T06:56:23Z)
• GPT4SGG: Synthesizing Scene Graphs from Holistic and Region-specific Narratives [69.36723767339001]
  We propose a novel framework named textitGPT4SGG to obtain more accurate and comprehensive scene graph signals. We show textitGPT4SGG significantly improves the performance of SGG models trained on image-caption data.
  arXiv  Detail & Related papers  (2023-12-07T14:11:00Z)
• Text-Only Training for Visual Storytelling [107.19873669536523]
  We formulate visual storytelling as a visual-conditioned story generation problem. We propose a text-only training method that separates the learning of cross-modality alignment and story generation.
  arXiv  Detail & Related papers  (2023-08-17T09:32:17Z)
• ViSTA: Vision and Scene Text Aggregation for Cross-Modal Retrieval [66.66400551173619]
  We propose a full transformer architecture to unify cross-modal retrieval scenarios in a single $textbfVi$sion. We develop dual contrastive learning losses to embed both image-text pairs and fusion-text pairs into a common cross-modal space. Experimental results show that ViSTA outperforms other methods by at least $bf8.4%$ at Recall@1 for scene text aware retrieval task.
  arXiv  Detail & Related papers  (2022-03-31T03:40:21Z)
• Global-Local Context Network for Person Search [125.51080862575326]
  Person search aims to jointly localize and identify a query person from natural, uncropped images. We exploit rich context information globally and locally surrounding the target person, which we refer to scene and group context, respectively. We propose a unified global-local context network (GLCNet) with the intuitive aim of feature enhancement.
  arXiv  Detail & Related papers  (2021-12-05T07:38:53Z)
• Text-based Person Search in Full Images via Semantic-Driven Proposal Generation [42.25611020956918]
  We propose a new end-to-end learning framework which jointly optimize the pedestrian detection, identification and visual-semantic feature embedding tasks. To take full advantage of the query text, the semantic features are leveraged to instruct the Region Proposal Network to pay more attention to the text-described proposals.
  arXiv  Detail & Related papers  (2021-09-27T11:42:40Z)
• StacMR: Scene-Text Aware Cross-Modal Retrieval [19.54677614738065]
  Cross-modal retrieval models have benefited from an increasingly rich understanding of visual scenes. Current models overlook a key aspect: the text appearing in images, which may contain crucial information for retrieval. We propose a new dataset that allows exploration of cross-modal retrieval where images contain scene-text instances.
  arXiv  Detail & Related papers  (2020-12-08T10:04:25Z)
• Probing Contextual Language Models for Common Ground with Visual Representations [76.05769268286038]
  We design a probing model that evaluates how effective are text-only representations in distinguishing between matching and non-matching visual representations. Our findings show that language representations alone provide a strong signal for retrieving image patches from the correct object categories. Visually grounded language models slightly outperform text-only language models in instance retrieval, but greatly under-perform humans.
  arXiv  Detail & Related papers  (2020-05-01T21:28:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.