Unsupervised contrastive learning achieves great success in learning image
representations with CNN. Unlike most recent methods that focused on improving
accuracy of image classification, we present a novel contrastive learning
approach, named DetCo, which fully explores the contrasts between global image
and local image patches to learn discriminative representations for object
detection. DetCo has several appealing benefits. (1) It is carefully designed
by investigating the weaknesses of current self-supervised methods, which
discard important representations for object detection. (2) DetCo builds
hierarchical intermediate contrastive losses between global image and local
patches to improve object detection, while maintaining global representations
for image recognition. Theoretical analysis shows that the local patches
actually remove the contextual information of an image, improving the lower
bound of mutual information for better contrastive learning. (3) Extensive
experiments on PASCAL VOC, COCO and Cityscapes demonstrate that DetCo not only
outperforms state-of-the-art methods on object detection, but also on
segmentation, pose estimation, and 3D shape prediction, while it is still
competitive on image classification. For example, on PASCAL VOC, DetCo-100ep
achieves 57.4 mAP, which is on par with the result of MoCov2-800ep. Moreover,
DetCo consistently outperforms supervised method by 1.6/1.2/1.0 AP on Mask
RCNN-C4/FPN/RetinaNet with 1x schedule. Code will be released at
\href{this https URL}{\color{blue}{\tt
github.com/xieenze/DetCo}}.

DetCo 是一种创新的对比学习方法，通过充分探索全局图像和局部图像补丁之间的对比，学习有助于目标检测的判别性表示，实验表明其不仅在目标检测方面优于现有方法，在分割，姿态估计和三维形状预测方面也显著优于监督方法。