Re-Scoring Using Image-Language Similarity for Few-Shot Object Detection

1 code implementation • 1 Nov 2023 • Min Jae Jung, Seung Dae Han, Joohee Kim

The former adapts CLIP, which performs zero-shot classification, to re-score the classification scores of a detector using image-class similarities, the latter is modified classification loss considering the punishment for fake backgrounds as well as confusing categories on a generalized few-shot object detection dataset.

Classification Few-Shot Object Detection +3