Abstract:
Motion, depth, and camera shake cause dynamic scene blur. Image segmentation or fully end-to-end trainable deep convolutional neural networks are used to solve this problem. Depth variations make these algorithms less effective.
We propose a depth-map-based deep neural convolutional network for dynamic scene deblurring. We extract the depth map from a blurred image and use a depth refinement network to restore edges and structure.
The spatial feature transform layer extracts depth features and merges them with image features through scaling and shifting to maximize depth map use. The depth map guides our image deblurring network to restore clarity.
We demonstrate that depth information is essential to the model’s performance through extensive experiments and analysis.
Finally, extensive quantitative and qualitative evaluations show that the proposed model outperforms state-of-the-art dynamic scene deblurring approaches and conventional depth-based algorithms.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
