paper-with-me

BigDetection

홈페이지 · 논문 7편

BigDetection is a new large-scale benchmark to build more general and powerful object detection systems. It leverages the training data from existing datasets ([LVIS](/dataset/lvis), [OpenImages](/dataset/openimages-v6) and [Object365](/dataset/objects365)) with carefully designed principles, and curate a larger dataset for improved detector pre-training. BigDetection dataset has 600 object categories and contains 3.4M training images with 36M object bounding boxes.

Images

벤치마크

Object Detection on BigDetection val 결과 30개