Scale-aware Automatic Augmentation for Object Detection
This work addresses the challenge of efficient and effective data augmentation for object detection, offering transferable policies that benefit tasks like instance segmentation, though it is incremental as it builds on existing automated augmentation approaches.
The authors tackled the problem of learning data augmentation policies for object detection by proposing Scale-aware AutoAug, which includes a scale-aware search space and a Pareto Scale Balance metric, resulting in significant and consistent improvements on various detectors like RetinaNet and Faster R-CNN, with much lower search costs than previous methods.
We propose Scale-aware AutoAug to learn data augmentation policies for object detection. We define a new scale-aware search space, where both image- and box-level augmentations are designed for maintaining scale invariance. Upon this search space, we propose a new search metric, termed Pareto Scale Balance, to facilitate search with high efficiency. In experiments, Scale-aware AutoAug yields significant and consistent improvement on various object detectors (e.g., RetinaNet, Faster R-CNN, Mask R-CNN, and FCOS), even compared with strong multi-scale training baselines. Our searched augmentation policies are transferable to other datasets and box-level tasks beyond object detection (e.g., instance segmentation and keypoint estimation) to improve performance. The search cost is much less than previous automated augmentation approaches for object detection. It is notable that our searched policies have meaningful patterns, which intuitively provide valuable insight for human data augmentation design. Code and models will be available at https://github.com/Jia-Research-Lab/SA-AutoAug.