From Web to Pixels: Bringing Agentic Search into Visual Perception

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Bokang, Sun, Xinyi, Feng, Kaituo, Dong, Xingping, Wu, Dongming, Yue, Xiangyu
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917487233204224
author Yang, Bokang
Sun, Xinyi
Feng, Kaituo
Dong, Xingping
Wu, Dongming
Yue, Xiangyu
author_facet Yang, Bokang
Sun, Xinyi
Feng, Kaituo
Dong, Xingping
Wu, Dongming
Yue, Xiangyu
contents Visual perception connects high-level semantic understanding to pixel-level perception, but most existing settings assume that the decisive evidence for identifying a target is already in the image or frozen model knowledge. We study a more practical yet harder open-world case where a visible object must first be resolved from external facts, recent events, long-tail entities, or multi-hop relations before it can be localized. We formalize this challenge as Perception Deep Research and introduce WebEye, an object-anchored benchmark with verifiable evidence, knowledge-intensive queries, precise box/mask annotations, and three task views: Search-based Grounding, Search-based Segmentation, and Search-based VQA. WebEyes contains 120 images, 473 annotated object instances, 645 unique QA pairs, and 1,927 task samples. We further propose Pixel-Searcher, an agentic search-to-pixel workflow that resolves hidden target identities and binds them to boxes, masks, or grounded answers. Experiments show that Pixel-Searcher achieves the strongest open-source performance across all three task views, while failures mainly arise from evidence acquisition, identity resolution, and visual instance binding.
format Preprint
id arxiv_https___arxiv_org_abs_2605_12497
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Web to Pixels: Bringing Agentic Search into Visual Perception
Yang, Bokang
Sun, Xinyi
Feng, Kaituo
Dong, Xingping
Wu, Dongming
Yue, Xiangyu
Computer Vision and Pattern Recognition
Visual perception connects high-level semantic understanding to pixel-level perception, but most existing settings assume that the decisive evidence for identifying a target is already in the image or frozen model knowledge. We study a more practical yet harder open-world case where a visible object must first be resolved from external facts, recent events, long-tail entities, or multi-hop relations before it can be localized. We formalize this challenge as Perception Deep Research and introduce WebEye, an object-anchored benchmark with verifiable evidence, knowledge-intensive queries, precise box/mask annotations, and three task views: Search-based Grounding, Search-based Segmentation, and Search-based VQA. WebEyes contains 120 images, 473 annotated object instances, 645 unique QA pairs, and 1,927 task samples. We further propose Pixel-Searcher, an agentic search-to-pixel workflow that resolves hidden target identities and binds them to boxes, masks, or grounded answers. Experiments show that Pixel-Searcher achieves the strongest open-source performance across all three task views, while failures mainly arise from evidence acquisition, identity resolution, and visual instance binding.
title From Web to Pixels: Bringing Agentic Search into Visual Perception
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.12497