https://ccdydy.github.io/tag/image-coding/Image CodingHugo Blox Builder (https://hugoblox.com)en-usMon, 26 May 2025 00:00:00 +0000https://ccdydy.github.io/media/icon_hu0b7a4cb9992c9ac0e91bd28ffd38dd00_9727_512x512_fill_lanczos_center_3.pnghttps://ccdydy.github.io/tag/image-coding/https://ccdydy.github.io/project/icmh/Mon, 26 May 2025 00:00:00 +0000https://ccdydy.github.io/project/icmh/<p>Images and videos are increasingly consumed not only by humans but also by recognition models. In collaboration with <strong>Yui Tatsumi</strong> (Watanabe Lab), this project studies coding frameworks that serve both machine vision and human viewing.</p> <ul> <li><strong>Explicit residual-based scalable coding</strong> (FR-ICMH / PR-ICMH) improves the coding efficiency and interpretability of scalable coding for humans and machines, with up to 29.57% BD-rate savings. <strong>(IEEE MMSP 2025)</strong></li> <li><strong>Seed selection</strong> picks the best random seed from early reverse-diffusion outputs, improving human-oriented reconstruction without extra bitrate. <strong>(IEEE GCCE 2025, Oral Presentation Award)</strong></li> <li><strong>Training-free adaptive quantization</strong> enables continuous variable-rate control for image coding for machines, with up to 11.07% BD-rate savings. <strong>(IEEE ICCE 2026)</strong></li> <li>Scalable coding for humans and machines extended to <strong>video via keyframe interpolation</strong>. <strong>(MIRU 2025; MIRU 2026, Interactive Presentation Award)</strong></li> </ul>