Frame-Level Fountain Coding for Low-Latency Real-Time Video
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fountain encoding techniques are unsuitable for real-time video applications due to excessive decoding delay, resulting in choppy, low-quality video.
Innovation Solution
Performing fountain encoding at the frame-level, where I-frames are encoded on a frame-by-frame basis, allowing for immediate display of decoded I-frames while still receiving and decoding P-frames and B-frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional fountain encoding is applied to entire data segments (GoP-level), then error recovery capability is improved, but decoding delay increases excessively
Solution Approach 1:
The patent divides the data segment (GoP) into individual frame-level units for independent fountain encoding. Each I-frame, P-frame, and B-frame is encoded separately rather than encoding the entire GoP as one unit. This segmentation allows the receiver to decode and display frames as they are received without waiting for the entire GoP to be transmitted and decoded, thus reducing decoding delay while maintaining error recovery capability at each frame level.
2Reliability
If fountain encoding is performed on complete GoP segments, then transmission reliability is improved, but video quality deteriorates due to choppy display
Solution Approach 1:
By segmenting the encoding unit from GoP-level to frame-level, the patent enables continuous video playback. The receiver can decode and display I-frames immediately upon receiving sufficient encoded packets, without waiting for the complete GoP. This eliminates the choppy video quality issue while maintaining transmission reliability through fountain coding's inherent error recovery capabilities at each frame level.
Solution Approach 2:
The patent applies fountain encoding to each frame independently in advance, allowing the receiver to have decoded frames ready for immediate display. This preliminary frame-level encoding preparation enables smooth video playback by ensuring that individual frames can be displayed as soon as they are decoded, without requiring the entire GoP to be processed first.
3Speed
If frame-level fountain encoding is used, then decoding speed is improved, but device complexity increases
Solution Approach 1:
While frame-level segmentation does increase encoding complexity compared to GoP-level encoding, the patent justifies this by demonstrating that it enables real-time video playback with reduced buffering requirements. The increased complexity is traded off for significantly improved decoding speed and video quality, making the system suitable for real-time applications where latency must be minimized.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Real-time video can be efficiently and reliably streamed over wireless networks by fountain encoding a stream of video packets at the frame-level. Specifically, deep packet inspection may be performed on the stream of video packets to identify a set of video packets corresponding to an intra-coded frame (I-frame). The set of video packet may then be fountain encoded together as a data segment to produce a set of fountain encoded packets corresponding to the I-frame. Video packets corresponding to other frames (e.g., P-frames, etc.) may also be encoded on a frame-by-frame basis. This may allow a user equipment (UE) to begin displaying earlier frames in a group of pictures (GoP) while fountain encoded packets for subsequent frames are still being received, thereby reducing latency and improving the overall user experience.