awesome-few-shot-object-detection icon indicating copy to clipboard operation
awesome-few-shot-object-detection copied to clipboard

Published 20 hours ago • verified publisher icon 2006pmach
→

Metadata

Awesome Few-Shot Object Detection Awesome

A (not complete) list of few-shot object detection methods for computer vision. While most methods include both, novel and base classes only a subset evaluates the performance separately for base and novel classes. The evaluation on base classes shows whether the method suffers from catastrophic forgetting or not.

Title Without Forgetting Few-Shot Training Venue
Generalized Few-Shot Object Detection without Forgetting Yes Fine-tuning CVPR'21
Beyond Max-Margin: Class Margin Equilibrium for Few-shot Object Detection No Meta-training CVPR'21
Dense Relation Distillation with Context-aware Aggregation for Few-Shot Object Detection No Meta-training CVPR'21
Transformation Invariant Few-Shot Object Detection No Meta-training CVPR'21
Few-Shot Object Detection via Classification Refinement and Distractor Retreatment No Meta-training CVPR'21
FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding Yes Fine-tuning CVPR'21
Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection Yes Fine-tuning CVPR'21
Accurate Few-shot Object Detection with Support-Query Mutual Guidance and Hybrid Loss No Meta-training CVPR'21
Hallucination Improves Few-Shot Object Detection Yes Fine-tuning CVPR'21
Meta Faster R-CNN: Towards Accurate Few-Shot Object Detection with Attentive Feature Alignment Yes Meta-training ArXiv'21
Meta-DETR: Few-Shot Object Detection via Unified Image-Level Meta-Learning Yes Meta-training ArXiv'21
Multi-Scale Positive Sample Refinement for Few-Shot Object Detection Yes Fine-tuning ECCV'20
Incremental Few-Shot Object Detection Yes Meta-training CVPR'20
Few-Shot Object Detection with Attention-RPN and Multi-Relation Detector No Meta-training CVPR'20
Frustratingly Simple Few-Shot Object Detection Yes Fine-tuning ICML'20
Meta R-CNN : Towards General Solver for Instance-level Few-shot Learning Yes Meta-training ICCV'19

← Metadata

17
Stars
1
Forks
Watchers

Owner

verified publisher icon 2006pmach

Metadata

Back