LiG-DETR: Local-in-Global Reassembly in Latent Space for Aerial Object Detection
To this end, we propose LiG-DETR, an Efficient Global-Local Reassembly framework that reformulates image slicing as high-fidelity local feature acquisition.
ProofPaper ↗
Key points
- Aerial object detection faces substantial scale and density variations.
- Existing methods mainly follow two paradigms: image slicing provides clearer local evidence but relies on independent crop-level prediction and post-processing, whereas feature- and query-level optimization preserves unified inference but operates on already compressed full-image representations, limiting recovery of fine-grained information.
- This raises a key question: can aerial detection directly acquire high-fidelity local evidence before feature degradation and integrate it into a unified end-to-end framework?
- A shared encoder extracts global and locally magnified features, which are projected into the detector feature space.
Sources (1)
- [1]LiG-DETR: Local-in-Global Reassembly in Latent Space for Aerial Object DetectionarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 06:07 AM
To this end, we propose LiG-DETR, an Efficient Global-Local Reassembly framework that reformulates image slicing as high-fidelity local feature acquisition.
Aerial object detection faces substantial scale and density variations.
Extractive summary: sentences quoted from the sources.