Meta-DETR: Few-Shot Object Detection via Unified Image-Level Meta-Learning
❓How?
Eliminate region-wise prediction and instead meta-learn object localization and classification at image level in a unified and complementary manner.
🛠️arxiv.org/abs/2103.11731
1/K ...👇
❓How?
Eliminate region-wise prediction and instead meta-learn object localization and classification at image level in a unified and complementary manner.
🛠️arxiv.org/abs/2103.11731
1/K ...👇
Specifically, the Meta-DETR first encodes both support and query images into category-specific
features and then feeds them into a category-agnostic decoder to directly generate predictions for specific categories. ...
2/K
features and then feeds them into a category-agnostic decoder to directly generate predictions for specific categories. ...
2/K
Authors propose a Semantic Alignment Mechanism (SAM), which aligns high-level and low-level feature semantics to improve the generalization of meta-learned representations. ...
3/K
3/K
➕SOTA Results for Few-shot detection on MS COCO.
The proposed Meta-DETR meta-learns object localization and classification at image level in a unified and complementary manner (without region-wise prediction), leading to superior few-shot object detection performance.
4/4
The proposed Meta-DETR meta-learns object localization and classification at image level in a unified and complementary manner (without region-wise prediction), leading to superior few-shot object detection performance.
4/4
@threadreaderapp unroll
• • •
Missing some Tweet in this thread? You can try to
force a refresh
