Open-Vocabulary Object Detection
Overview
Open-vocabulary object detection studies how to find and localize objects described by language, including categories that were not labeled during detector training. Common approaches align image regions with text using vision-language models such as CLIP, then improve localization and recognition through grounding, prompting, pseudo-labeling, or adaptation. Methods are commonly evaluated on benchmarks such as COCO and LVIS, with growing attention to fine-grained queries, domain shifts, and open-world settings where unknown objects may appear. A central challenge is making predictions reliable across wording and domains while calibrating confidence so that novel objects are not confused with background or familiar categories.
Latest papers
94 papers5 papers in the last four weeks, down 44% on the four weeks before. 0.0% of all new papers.
Generalized Decoupled Learning for Enhancing Open-Vocabulary Dense Perception
content'' and context'' features respectively. \revise{The context features are enhanced by jointly distilling semantic correlations from Vision Foundation Models (VFMs) and object integrity cues from diffusion models, thereby enhancing spatial consistency. In parallel, the content features are aligned with image crop representations and constrained by region correlations from VFMs to improve local discriminability. Extensive experiments demonstrate that DeCLIP establishes a solid foundation for open-vocabulary dense perception, consistently achieving state-of-the-art performance across a broad spectrum of tasks, including 2D detection and segmentation, 3D instance segmentation, video instance segmentation, and 6D object pose estimation.} Code is available at https://github.com/xiaomoguhz/DeCLIP