CVPR Findings2026
OV-Stitcher: A Global Context-Aware Framework for Training-Free Open Vocabulary Semantic Segmentation
Seungjae Moon, Seunghyun Oh, Youngmin Ro
Abstract
Training-free open-vocabulary semantic segmentation (TF-OVSS) has recently attracted attention for its ability to perform dense prediction by leveraging the pretrained knowledge of large vision and vision-language models, without requiring additional training. However, due to the limited input resolution of these pretrained encoders, existing TF-OVSS methods commonly adopt a sliding-window strategy that processes cropped sub-images independently. While effective for managing high-resolution inputs, this approach prevents global attention over the full image, leading to fragmented feature representations and limited contextual reasoning. We propose OV-Stitcher, a training-free framework that addresses this limitation by stitching fragmented sub-image features directly within the final encoder block. By reconstructing attention representations from fragmented sub-image features, OV-Stitcher enables global attention within the final encoder block, producing coherent context aggregation and spatially consistent, semantically aligned segmentation maps. Extensive evaluations across eight benchmarks demonstrate that OV-Stitcher establishes a scalable and effective solution for open-vocabulary segmentation, achieving a notable improvement in mean Intersection over Union (mIoU) from 48.7 to 50.7 compared with prior training-free baselines.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on40
- Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
- Segment AnythingAlexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao et al.ICCV 2023 · 13,211 citations
- Emerging Properties in Self-Supervised Vision TransformersMathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou et al.ICCV 2021 · 8,921 citations
- Sigmoid Loss for Language Image Pre-TrainingXiaohua Zhai, Basil Mustafa, Alexander Kolesnikov, Lucas BeyerICCV 2023 · 2,932 citations
- FlashAttention-2: Faster Attention with Better Parallelism and Work PartitioningTri DaoICLR 2024 · 2,600 citations
Related papers
- Training-free Open-Vocabulary Semantic Segmentation via Diverse Prototype Construction and Sub-region MatchingXuanpu Zhao, Dianmo Sheng, Zhentao Tan, Zhiwei Zhao et al.AAAI 2025 · 2 citations
- Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic SegmentationChanyoung Kim, Dayun Ju, Woojung Han, Ming-Hsuan Yang et al.CVPR 2025
- Emergent Open-Vocabulary Semantic Segmentation from Off-the-Shelf Vision-Language ModelsJiayun Luo, Siddhesh Khandelwal, Leonid Sigal, Boyang LiCVPR 2024 · 10 citations
- LPOSS: Label Propagation Over Patches and Pixels for Open-vocabulary Semantic SegmentationVladan Stojnic, Yannis Kalantidis, Jirí Matas, Giorgos ToliasCVPR 2025
- Training-Free Class Purification for Open-Vocabulary Semantic SegmentationQi Chen, Lingxiao Yang, Yun Chen, Nailong Zhao et al.ICCV 2025 · 4 citations
