GAIA: A Transfer Learning System of Object Detection that Fits Your Needs
This work addresses the challenge for object detection practitioners who require efficient, task-specific models without extensive manual tuning, though it is incremental as it builds on existing transfer learning paradigms.
The paper tackles the problem of efficiently customizing object detection models for diverse downstream needs like latency constraints and specialized data distributions, presenting GAIA, a transfer learning system that automatically provides tailored solutions and achieves AP scores from 38.2 to 46.5 on COCO with latencies ranging from 16ms to 53ms.
Transfer learning with pre-training on large-scale datasets has played an increasingly significant role in computer vision and natural language processing recently. However, as there exist numerous application scenarios that have distinctive demands such as certain latency constraints and specialized data distributions, it is prohibitively expensive to take advantage of large-scale pre-training for per-task requirements. In this paper, we focus on the area of object detection and present a transfer learning system named GAIA, which could automatically and efficiently give birth to customized solutions according to heterogeneous downstream needs. GAIA is capable of providing powerful pre-trained weights, selecting models that conform to downstream demands such as latency constraints and specified data domains, and collecting relevant data for practitioners who have very few datapoints for their tasks. With GAIA, we achieve promising results on COCO, Objects365, Open Images, Caltech, CityPersons, and UODB which is a collection of datasets including KITTI, VOC, WiderFace, DOTA, Clipart, Comic, and more. Taking COCO as an example, GAIA is able to efficiently produce models covering a wide range of latency from 16ms to 53ms, and yields AP from 38.2 to 46.5 without whistles and bells. To benefit every practitioner in the community of object detection, GAIA is released at https://github.com/GAIA-vision.