GazeAnywhere：基于概念提示的任意场景凝视目标估计

AI 摘要

研究人员提出了一种新的可提示凝视目标估计（PGE）任务，通过文本或视觉提示识别主体并预测凝视目标，无需多阶段流水线。他们开发了Gaze-Co数据集（包含12万张带提示标注的图像对）和GazeAnywhere模型，该模型使用基于Transformer的检测器融合冻结编码器特征，同时解决主体定位、帧内/外存在性和凝视目标热图估计。GazeAnywhere在多个PGE基准上达到最先进性能，并已在GitHub上开源，论文被CVPR 2026接收。

正文节选

Gaze Target Estimation Anywhere with Concepts Abstract A new promptable paradigm integrates subject localization and gaze estimation into an end-to-end transformer model that uses text or visual prompts to identify subjects and predict gaze targets without multi-stage pipelines. Estimating human gaze targets from images in-the-wild is an important and formidable task. Existing approaches primarily employ brittle, multi-stage pipelines that require explicit inputs, like head bounding boxes and hu

发布时间：—

抓取时间：2026-08-14 05:08

来源机构：Hugging Face