| Paper ID | CI-4.3 |
| Paper Title |
LEARNING TO ESTIMATE KERNEL SCALE AND ORIENTATION OF DEFOCUS BLUR WITH ASYMMETRIC CODED APERTURE |
| Authors |
Jisheng Li, Tsinghua University, China; Qi Dai, Boyan Technology DBA RayShaper China, China; Jiangtao Wen, Tsinghua University, China |
| Session | CI-4: Remote Sensing and Coded Aperture Imaging |
| Location | Gather.Town |
| Session Time: | Thursday, 10 June, 15:30 - 16:15 |
| Presentation Time: | Thursday, 10 June, 15:30 - 16:15 |
| Presentation |
Poster
|
| Topic |
Computational Imaging: [IMT] Computational Imaging Methods and Models |
| IEEE Xplore Open Preview |
Click here to view in IEEE Xplore |
| Virtual Presentation |
Click here to watch in the Virtual Conference |
| Abstract |
Consistent in-focus input imagery is an essential precondition for machine vision systems to perceive the dynamic environment. A defocus blur severely degrades the performance of vision systems. To tackle this problem, we propose a deep-learning-based framework estimating the kernel scale and orientation of the defocus blur to adjust lens focus rapidly. Our pipeline utilizes 3D ConvNet for a variable number of input hypotheses to select the optimal slice from the input stack. We use random shuffle and Gumbel-softmax to improve network performance. We also propose to generate synthetic defocused images with various asymmetric coded apertures to facilitate training. Experiments are conducted to demonstrate the effectiveness of our framework. |
