SIEDD: Shared-Implicit Encoder with Discrete Decoders
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918075211710464 |
|---|---|
| author | Rangarajan, Vikram Maiya, Shishira Ehrlich, Max Shrivastava, Abhinav |
| author_facet | Rangarajan, Vikram Maiya, Shishira Ehrlich, Max Shrivastava, Abhinav |
| contents | Implicit Neural Representations (INRs) offer exceptional fidelity for video compression by learning per-video optimized functions, but their adoption is crippled by impractically slow encoding times. Existing attempts to accelerate INR encoding often sacrifice reconstruction quality or crucial coordinate-level control essential for adaptive streaming and transcoding. We introduce SIEDD (Shared-Implicit Encoder with Discrete Decoders), a novel architecture that fundamentally accelerates INR encoding without these compromises. SIEDD first rapidly trains a shared, coordinate-based encoder on sparse anchor frames to efficiently capture global, low-frequency video features. This encoder is then frozen, enabling massively parallel training of lightweight, discrete decoders for individual frame groups, further expedited by aggressive coordinate-space sampling. This synergistic design delivers a remarkable 20-30X encoding speed-up over state-of-the-art INR codecs on HD and 4K benchmarks, while maintaining competitive reconstruction quality and compression ratios. Critically, SIEDD retains full coordinate-based control, enabling continuous resolution decoding and eliminating costly transcoding. Our approach significantly advances the practicality of high-fidelity neural video compression, demonstrating a scalable and efficient path towards real-world deployment. Our codebase is available at https://github.com/VikramRangarajan/SIEDD . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_23382 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SIEDD: Shared-Implicit Encoder with Discrete Decoders Rangarajan, Vikram Maiya, Shishira Ehrlich, Max Shrivastava, Abhinav Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Implicit Neural Representations (INRs) offer exceptional fidelity for video compression by learning per-video optimized functions, but their adoption is crippled by impractically slow encoding times. Existing attempts to accelerate INR encoding often sacrifice reconstruction quality or crucial coordinate-level control essential for adaptive streaming and transcoding. We introduce SIEDD (Shared-Implicit Encoder with Discrete Decoders), a novel architecture that fundamentally accelerates INR encoding without these compromises. SIEDD first rapidly trains a shared, coordinate-based encoder on sparse anchor frames to efficiently capture global, low-frequency video features. This encoder is then frozen, enabling massively parallel training of lightweight, discrete decoders for individual frame groups, further expedited by aggressive coordinate-space sampling. This synergistic design delivers a remarkable 20-30X encoding speed-up over state-of-the-art INR codecs on HD and 4K benchmarks, while maintaining competitive reconstruction quality and compression ratios. Critically, SIEDD retains full coordinate-based control, enabling continuous resolution decoding and eliminating costly transcoding. Our approach significantly advances the practicality of high-fidelity neural video compression, demonstrating a scalable and efficient path towards real-world deployment. Our codebase is available at https://github.com/VikramRangarajan/SIEDD . |
| title | SIEDD: Shared-Implicit Encoder with Discrete Decoders |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2506.23382 |