jigsaw-puzzle
Jigsaw-Puzzles
Jigsaw-Puzzles Dataset
Jigsaw-Puzzles is a novel benchmark consisting of 1,100 carefully curated real-world images with high spatial complexity, designed to rigorously evaluate Vision-Language Models' (VLMs) spatial perception, structural understanding, and reasoning capabilities. The dataset minimizes reliance on domain-specific knowledge to better isolate and assess general spatial reasoning, positioning itself as a challenging and diagnostic benchmark for advancing spatial… See the full description on the dataset page: https://huggingface.co/datasets/zesen01/Jigsaw-Puzzles.jigsaw_puzzleImages of various jigsaw puzzle pieces gathered from a website called Jigsaw Explorer.
Segmentations were generated by SAM 3 with some image preprocessing. Details about how I created this dataset can be found in this github repository
Follows the schema used by YOLO segmentation models, images are resized to 1920x1080
imnet1k_jigsaw_puzzle
