Can we learn how to detect an object if we just have one label? This is the key question that we are trying to solve in DARPA Learning with Less Labels program. During the first year of the program, we performed well on an international benchmark.

Other teams are key players in AI in USA (e.g., Berkeley) and Australia. It is very nice to collaborate with and learn from these other teams.

The tasks include: image classification, object detection and video classification. Due to our performance in Phase 1, we have been selected for Phase 2 of the program. We are very proud - looking forward to improve further on few-shot learning!

This research is sponsored by DARPA Learning with Less Labels.