Self-supervised Learning with Local Contrastive Loss for Detection and Semantic Segmentation
This work addresses the need for better self-supervised methods in computer vision tasks like detection and segmentation, offering incremental improvements over existing approaches.
The paper tackles the problem of improving self-supervised learning for object detection and semantic segmentation by introducing a local contrastive loss that enforces pixel-level consistency, resulting in performance gains of 1.9% on COCO detection, 1.4% on PASCAL VOC detection, and 0.6% on CityScapes segmentation.
We present a self-supervised learning (SSL) method suitable for semi-global tasks such as object detection and semantic segmentation. We enforce local consistency between self-learned features, representing corresponding image locations of transformed versions of the same image, by minimizing a pixel-level local contrastive (LC) loss during training. LC-loss can be added to existing self-supervised learning methods with minimal overhead. We evaluate our SSL approach on two downstream tasks -- object detection and semantic segmentation, using COCO, PASCAL VOC, and CityScapes datasets. Our method outperforms the existing state-of-the-art SSL approaches by 1.9% on COCO object detection, 1.4% on PASCAL VOC detection, and 0.6% on CityScapes segmentation.