Research map / Architectures and efficiency / Quantization
LVQAC: Lattice Vector Quantization Coupled With Spatially Adaptive Companding for Efficient Learned Image Compression
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Replace scalar quantization with LVQAC | Cheng2020 with autoregressive context · Kodak | BD-Rate: 3.35 -> 3.12 | Helped |
| Replace scalar quantization with LVQAC | Cheng2020 with checkerboard context · Kodak | BD-Rate: 3.89 -> 3.39 | Helped |
| Replace scalar quantization with LVQAC | Minnen2018 with autoregressive context · Kodak | BD-Rate: 14.92 -> 14.75 | Helped |
| Replace scalar quantization with LVQAC | Minnen2018 with checkerboard context · Kodak | BD-Rate: 20.00 -> 16.17 | Helped |
| Add spatially adaptive companding to LVQ (ablation) | Cheng2020 with autoregressive context · Kodak | BD-Rate: 3.24 -> 3.12 | Helped |
| Add LVQ to scalar quantization with AC (ablation) | Cheng2020 with autoregressive context · Kodak | BD-Rate: 3.30 -> 3.12 | Helped |
| Add spatially adaptive companding to LVQ (ablation) | Cheng2020 with checkerboard context · Kodak | BD-Rate: 3.68 -> 3.39 | Helped |
| Add LVQ to scalar quantization with AC (ablation) | Cheng2020 with checkerboard context · Kodak | BD-Rate: 3.82 -> 3.39 | Helped |
| Add spatially adaptive companding to LVQ (ablation) | Minnen2018 with autoregressive context · Kodak | BD-Rate: 14.86 -> 14.75 | Helped |
| Add LVQ to scalar quantization with AC (ablation) | Minnen2018 with autoregressive context · Kodak | BD-Rate: 14.88 -> 14.75 | Helped |
| Add spatially adaptive companding to LVQ (ablation) | Minnen2018 with checkerboard context · Kodak | BD-Rate: 18.43 -> 16.17 | Helped |
| Add LVQ to scalar quantization with AC (ablation) | Minnen2018 with checkerboard context · Kodak | BD-Rate: 19.29 -> 16.17 | Helped |
Closest papers by meaning
- Multirate Neural Image Compression with Adaptive Lattice Vector Quantization (CVPR 2025 · similarity 0.93)
- Learning Optimal Lattice Vector Quantizers for End-to-end Neural Image Compression (NeurIPS 2024 · similarity 0.93)
- Approaching Rate-Distortion Limits in Neural Compression with Lattice Transform Coding (ICLR 2025 · similarity 0.89)
- HyperVQ: Enabling Hyperprior Entropy Modeling for VQ-Based Generative Image Compression (NeurIPS 2026 · similarity 0.86)
- Binary Quadratic Quantization: Beyond First-Order Quantization for Real-Valued Matrix Compression (NeurIPS 2025 · similarity 0.86)
- Learning Grouped Lattice Vector Quantizers for Low-Bit LLM Compression (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.