WeVisOne-VL：从对象到像素的统一视觉感知模型
WeVisOne-VL: Unified Visual Perception from Objects to Pixels
WeVisOne-VL 通过结构化 token 预测统一对象级和像素级视觉感知，支持检测、定位、计数、OCR、分割和 matting 等任务，采用轻量级像素解码器和多阶段训练，在 30 项以上基准测试中表现领先，并保持对不存在对象的稳健拒绝。
同一件事另外 1 家也报道了
各家口径不同：Hugging Face Daily Papers 表示使用轻量级像素解码器；MarkTechPost 提到 9.6M 参数解码器。
- Hugging Face Daily Papers··首发·当前阅读
WeVisOne-VL：从对象到像素的统一视觉感知模型
- MarkTechPost··+12 天
腾讯 WeVisOne-VL 统一检测、OCR、分割和 Matting
新增信息：4B 版本在 Dense200 密集检测上达到 71.3 F1 均值。
按发布时间排列 · 事件后续出现时自动追加 · 由模型根据各家标题与摘要比对
Published on Oct 22
Authors:
,
,
,
,
,
,
Abstract
WeVisOne-VL unifies object-level and pixel-level visual perception via structured token prediction, supporting detection, localization, counting, OCR, segmentation, and matting with a lightweight pixel decoder and multi-stage training.
Existing vision-language models (VLMs) remain unreliable for fine-grained visual perception. In this report, we present WeVisOne-VL (WeChat Vision All-in-One), a powerful VLM that unifies object-level and pixel-level perception through structured next-token prediction. It supports multiple visual tasks including object detection, referring localization, counting, OCR, segmentation, and matting, enabling localization at any user-specific granularity, from points and boxes to masks and alpha mattes. Architecturally, we integrate a lightweight pixel decoder into the VLM via a special <OBJ> token to generate instance-level pixel outputs. We train the model with a three-stage curriculum SFT followed by mixed-task RL. Across over 30 benchmarks, WeVisOne-VL demonstrates leading performance on both object-level and pixel-level tasks, while maintaining robust rejection for object-absent queries. By releasing WeVisOne-VL to the open-source community, we hope to advance VLMs from broad visual understanding toward truly reliable perception systems for real-world deployment.
View arXiv page View PDF Project page GitHub 44 Add to collection
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.31329 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.31329 in a dataset README.md to link it from this page.
Spaces citing this paper 1
Collections including this paper 1
来源: Hugging Face Daily Papers · huggingface.co