AI 中文总结
针对AIGC图像超低比特率压缩，提出基于AEIC一步扩散编解码器的速率自适应系统，通过多检查点候选生成与背包选择，在0.025 BPP下实现高感知质量，DISTS排名第二。
AI 中文摘要
我们描述了我们参加LoViF 2026 AIGC图像压缩挑战赛的参赛方案，该挑战赛是在严格的总速率预算为每像素0.025比特（BPP）下对AI生成图像进行超低比特率编码的基准。生成的图像对压缩提出了独特的挑战：它们经常包含渲染的排版、合成边缘、重复的图案、类似UI的布局和风格化的微观纹理，而传统的面向失真的编解码器在此速率下会抹去这些内容，而无约束的生成解码器可以恢复看似合理的细节，但这些细节不再匹配源几何或符号。我们将此视为一个速率感知分配问题。我们的系统微调了AEIC一步扩散编解码器的四个速率专用检查点，从所有四个检查点生成每个图像的候选，包括一个通过编码器侧测试时优化（TTO）并定期进行熵条件刷新的潜在表示，使用实用的rANS编码对每个候选进行熵编码，并通过在真实编码文件大小上求解精确的多选背包问题，为每个图像选择一个比特流。在解码时应用一个固定的、零额外比特的残差恢复网络。每个提交的比特流都可以由随附的解码器独立解码，该解码器不使用源图像或外部辅助信息。我们报告了完整的流程、涵盖77次记录实验的消融历史，以及一组负面结果，包括为什么在这种架构下PSNR无法与面向速率失真的竞争对手持平，这对未来的参与者很有用。这是一份挑战报告：我们的参赛得分31.527739（PSNR 27.02 dB，MS-SSIM 0.9176，LPIPS 0.0778，DISTS 0.0390，在0.02495 BPP下），在排行榜上DISTS排名第二，在2026年8月4日公布的最终测试阶段排行榜上排名第五。
英文摘要
We describe our entry to the LoViF 2026 AIGC Image Compression Challenge, a benchmark for ultra-low-bitrate coding of AI-generated images under a strict global rate budget of 0.025 bits per pixel (BPP). Generated imagery poses a distinct challenge for compression: it frequently contains rendered typography, synthetic edges, repeated motifs, UI-like layout, and stylized micro-texture that conventional distortion-oriented codecs erase at this rate, while unconstrained generative decoders can restore plausible-looking detail that no longer matches the source geometry or symbols. We treat this as a rate-perception allocation problem. Our system fine-tunes four rate-specialized checkpoints of the AEIC one-step diffusion codec, generates per-image candidates from all four, including one latent refined through encoder-side test-time optimization (TTO) with periodic entropy-conditioning refresh, entropy-codes every candidate with practical rANS coding, and selects exactly one bitstream per image with an exact multiple-choice knapsack solved over true coded file sizes. A fixed, zero-additional-bit residual restoration network is applied at decode time. Every submitted bitstream is independently decodable by the shipped decoder, which uses no source image or external side information. We report the full pipeline, an ablation history spanning 77 logged experiments, and a set of negative results, including why PSNR could not be pushed to parity with rate-distortion-oriented competitors under this architecture, useful to future participants. This is a challenge report: our entry scored 31.527739 (PSNR 27.02 dB, MS-SSIM 0.9176, LPIPS 0.0778, DISTS 0.0390 at 0.02495 BPP), the second-best DISTS on the leaderboard, ranking 5th on the final test-phase leaderboard announced August 4, 2026.
