Class-Incremental Few-Shot Object Detection
It addresses a realistic challenge in object detection for AI systems, enabling incremental learning with limited data, though it is incremental in method.
The paper tackles the problem of class-incremental few-shot object detection, where models must learn new objects from few examples without forgetting previous ones, and proposes LEAST, a method that reduces forgetting and improves transfer capability, achieving significant performance gains on benchmarks.
Conventional detection networks usually need abundant labeled training samples, while humans can learn new concepts incrementally with just a few examples. This paper focuses on a more challenging but realistic class-incremental few-shot object detection problem (iFSD). It aims to incrementally transfer the model for novel objects from only a few annotated samples without catastrophically forgetting the previously learned ones. To tackle this problem, we propose a novel method LEAST, which can transfer with Less forgetting, fEwer training resources, And Stronger Transfer capability. Specifically, we first present the transfer strategy to reduce unnecessary weight adaptation and improve the transfer capability for iFSD. On this basis, we then integrate the knowledge distillation technique using a less resource-consuming approach to alleviate forgetting and propose a novel clustering-based exemplar selection process to preserve more discriminative features previously learned. Being a generic and effective method, LEAST can largely improve the iFSD performance on various benchmarks.