SOCO: Benchmarking Semantic Object Correspondence in Vision Foundation Models
Abstract
Semantic Object Correspondence (SOCO) benchmark evaluates structured object understanding in vision models through consistent part-level annotations and keypoint descriptions, revealing gaps between language-grounded localization and visual correspondence while demonstrating strong prediction of downstream task performance.
Measuring structured object understanding in vision foundation models remains challenging due to inconsistent evaluation protocols and limited part-level supervision. Semantic correspondence (SC) evaluates this capability by testing whether object parts can be matched across instances and categories under large variations in appearance, viewpoint, and geometry. To enable a systematic SC evaluation, we introduce SOCO, a new benchmark for Semantic Object Correspondence that introduces a taxonomy of correspondence types and provides consistent, functionally meaningful keypoint annotations across 100 categories and over 1M correspondence pairs. In addition, SOCO includes keypoint language descriptions, enabling the evaluation of large vision-language models (LVLMs) and their fine-grained part-level understanding. Comprehensive experiments reveal that (i) vision foundation backbones encode strong semantic structure but transfer correspondences poorly across related categories and only partially capture object-part position, (ii) LVLMs are stronger at text-prompted part localization than at visual-reference cross-image matching, exposing a gap between language-grounded localization and fine-grained visual correspondence, and (iii) correspondence performance predicts performance on dense downstream tasks, including segmentation, tracking, 3D pose estimation, and 3D detection, more strongly than ImageNet classification. Together, these findings position SOCO as a benchmark for structured, part-level representation quality in vision and multimodal foundation models.
Community
SOCO is a benchmark for evaluating structured, part-level understanding in vision and multimodal foundation models through semantic correspondence. It provides a taxonomy of correspondence types, functionally meaningful keypoint annotations across 100 categories, over 1M correspondence pairs, and language descriptions for evaluating LVLMs. Experiments show that current vision backbones encode semantic structure but struggle with cross-category correspondence and object-part position, while LVLMs perform better at text-prompted localization than visual-reference matching. SOCO also shows that semantic correspondence is a stronger predictor of dense downstream performance than ImageNet classification.
Get this paper in your agent:
hf papers read 2605.31597 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 1
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper