Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

AirDet: Few-Shot Detection without Fine-tuning for Autonomous Exploration

Dec 03, 2021
Bowen Li, Chen Wang, Pranay Reddy, Seungchan Kim, Sebastian Scherer

Share this with someone who'll enjoy it:

Few-shot object detection has rapidly progressed owing to the success of meta-learning strategies. However, the requirement of a fine-tuning stage in existing methods is timeconsuming and significantly hinders their usage in real-time applications such as autonomous exploration of low-power robots. To solve this problem, we present a brand new architecture, AirDet, which is free of fine-tuning by learning class agnostic relation with support images. Specifically, we propose a support-guided cross-scale (SCS) feature fusion network to generate object proposals, a global-local relation network (GLR) for shots aggregation, and a relation-based prototype embedding network (R-PEN) for precise localization. Exhaustive experiments are conducted on COCO and PASCAL VOC datasets, where surprisingly, AirDet achieves comparable or even better results than the exhaustively finetuned methods, reaching up to 40-60% improvements on the baseline. To our excitement, AirDet obtains favorable performance on multi-scale objects, especially the small ones. Furthermore, we present evaluation results on real-world exploration tests from the DARPA Subterranean Challenge, which strongly validate the feasibility of AirDet in robotics. The source code, pre-trained models, along with the real world data for exploration, will be made public.

* 10 pages, 8 figures 

   Access Paper Source

Share this with someone who'll enjoy it: