One-Shot Object Detection with Co-Attention and Co-Excitation
arXiv:1911.12529
Abstract
This paper aims to tackle the challenging problem of one-shot object detection. Given a query image patch whose class label is not included in the training data, the goal of the task is to detect all instances of the same class in a target image. To this end, we develop a novel {\em co-attention and co-excitation} (CoAE) framework that makes contributions in three key technical aspects. First, we propose to use the non-local operation to explore the co-attention embodied in each query-target pair and yield region proposals accounting for the one-shot situation. Second, we formulate a squeeze-and-co-excitation scheme that can adaptively emphasize correlated feature channels to help uncover relevant proposals and eventually the target objects. Third, we design a margin-based ranking loss for implicitly learning a metric to predict the similarity of a region proposal to the underlying query, no matter its class label is seen or unseen in training. The resulting model is therefore a two-stage detector that yields a strong baseline on both VOC and MS-COCO under one-shot setting of detecting objects from both seen and never-seen classes. Codes are available at https://github.com/timy90022/One-Shot-Object-Detection.
arXiv commentNeurIPS 2019
Cited by in corpus (15)
- A Survey on Visual Transformer
- UP-DETR: Unsupervised Pre-training for Object Detection with Transformers
- BriNet: Towards Bridging the Intra-class and Inter-class Gaps in One-Shot Segmentation
- Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection
- Cooperating RPN's Improve Few-Shot Object Detection
- Dynamic Relevance Learning for Few-Shot Object Detection
- Dense Relation Distillation with Context-aware Aggregation for Few-Shot Object Detection
- CoVA: Context-aware Visual Attention for Webpage Information Extraction
- Hallucination Improves Few-Shot Object Detection
- CAT: Cross-Attention Transformer for One-Shot Object Detection
- Leveraging Bottom-Up and Top-Down Attention for Few-Shot Object Detection
- Localizing Infinity-shaped fishes: Sketch-guided object localization in the wild
- Compositional Embeddings for Multi-Label One-Shot Learning
- Top-Related Meta-Learning Method for Few-Shot Object Detection
- SBNet: Segmentation-based Network for Natural Language-based Vehicle Search