Researchers have introduced a new framework and dataset for addressing heterogeneous stereo deblurring, a problem arising from hardware variations in smartphone cameras that cause asymmetric blur. The proposed physically- and epipolar-constrained cross attention (PECA) module enhances feature fusion by adhering to physically valid disparity constraints and employing a confidence-weighted attention mechanism. This approach improves deblurring performance across various architectures like CNNs, Transformers, and NAFNet, demonstrating efficiency and reliability. AI
IMPACT This research could lead to improved image quality in mobile XR applications by addressing specific hardware-related image degradation.
RANK_REASON The cluster contains an academic paper detailing a new benchmark, dataset, and method for a computer vision task.
- arXiv
- CNN
- NAFNet
- PECA
- Transformer
- alphaXiv
- CatalyzeX
- DagsHub
- Gotit.pub
- HSD dataset
- Hugging Face
- Influence Flower
- ScienceCast
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →