We are glad to announce that our paper, “Bound-Conditioned Latent Inference for Progressive Image Compression” (BLI), has been accepted to NeurIPS 2026!

BLI continues our lab’s research on progressive image compression, following DPICT (CVPR 2022) and CTC (CVPR 2023). This series explores how a single compressed bitstream can support a wide range of bitrates, progressively improving image quality as more bits are received. Starting with trit-plane coding in DPICT and advancing to context-based coding in CTC, we now take another step forward with BLI.

1. Jaeseok Jang, Seungmin Jeon, Kwang Pyo Choi, and Chang-Su Kim, “Bound-Conditioned Latent Inference for Progressive Image Compression,” NeurIPS 2026. (BLI)
2. Seungmin Jeon, Kwang Pyo Choi, Youngo Park, and Chang-Su Kim, “Context-Based Trit-Plane Coding for Progressive Image Compression,” CVPR 2023. (CTC)
3. Jae-Han Lee, Seungmin Jeon, Kwang Pyo Choi, Youngo Park, and Chang-Su Kim, “DPICT: Deep Progressive Image Compression Using Trit-Planes,” CVPR 2022. (DPICT)

Congratulations to Jaeseok and Seungmin! Our images improve progressively, and so does our research.