Video Image Pruning with Edge-Directed Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding techniques face challenges in reducing bit-rate while maintaining high video quality, as methods like increasing quantization step size result in artifacts, and preprocessing techniques like data pruning can make decoded videos appear blurry due to loss of high-frequency information.
Innovation Solution
The proposed method involves pre-processing video images by selectively removing portions in the spatial domain, such as diagonal lines, prior to compression, and using edge-directed interpolation to restore the original image size after decoding, allowing for flexible bit-rate reduction without filtering pixels, thus preserving video quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data pruning is performed to reduce bit-rate, then bit-rate is reduced, but video quality deteriorates due to loss of high frequency information
Solution Approach 1:
The patent applies preliminary action by performing data pruning before compression to reduce the amount of data that needs to be encoded. By selectively removing less important data (such as diagonal line portions) before the compression stage, the system reduces the overall bit-rate requirement while preserving the most critical visual information for quality maintenance.
Solution Approach 2:
The patent implements local quality by differentiating between different types of data in the video image. Instead of uniformly processing all data, the system identifies and treats different regions differently - specifically removing diagonal line portions while preserving other important data structures. This selective approach maintains local visual quality in regions that need it most while achieving overall bit-rate reduction.
2Quantity of substance
If low-pass filters and down-sampling are used to reduce bit-rate, then bit-rate is reduced, but decoded video appears blurry
Solution Approach 1:
The patent applies the extraction principle by selectively removing specific problematic data elements (diagonal line portions) from the image before compression. This targeted extraction differs from conventional low-pass filtering by removing only the specific data structures that cause blurriness while preserving the overall sharpness and detail of the decoded video.
Solution Approach 2:
Instead of using conventional filtering approaches that blur the entire image to reduce bit-rate, the patent inverts the approach by selectively removing only the problematic diagonal line portions. This inversion of the conventional wisdom allows the system to reduce bit-rate without the side effect of image blurriness.
3Area of stationary object
If conventional interpolation methods are used to restore image size, then image size is restored, but edges appear jerky
Solution Approach 1:
The patent applies local quality by using edge-directed interpolation that adapts to the local characteristics of the image. Instead of applying a uniform interpolation method across the entire image, the system analyzes edge information and adjusts the interpolation process locally to match the actual image structure. This ensures smooth edges are preserved while maintaining the restored image size.
Solution Approach 2:
The patent implements feedback by using the decoded image information to guide the interpolation process. The edge-directed interpolation method incorporates feedback about the image's edge structure to adjust the reconstruction process, ensuring that interpolated regions blend smoothly with existing edges rather than creating jerky transitions.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Methods and apparatus are provided for video image pruning. An apparatus includes a data pruner (110) for pre-processing a picture prior to, and in preparation for, compression by encoding. The data pruner (110) selectively removes, in the spatial domain, at least one region within the picture. At the decoder end, an apparatus includes a a data restorer (140) for receiving a decompressed picture subsequent to decompression by decoding, and post-processing the decompressed picture by selectively restoring, in the spatial domain, at least one region in the decompressed picture based on information indicating a removal of the at least one region prior to a previously performed encoding process.