Sampling Innovation-Based Adaptive Compressive Sensing

Zhifu Tian, Tao Hu, Chaoyang Niu, Di Wu, Shu Wang · CVPR 2025 · 2025

DOI arXiv

Status

To read

Topics

budgeted-inference, other

Feeds

P2 · NeurIPS 2027

What they did

For adaptive compressive sensing of images, predicts the "sampling innovation" (how much extra measurements would reduce reconstruction error) and allocates more sampling to regions that are hard to reconstruct.

My note

Allocates image measurements using predicted reconstruction-error reduction from extra samples. The strongest cross-domain novelty check.

BibTeX

@inproceedings{tian2025-sib-acs,
  title = {{Sampling Innovation-Based Adaptive Compressive Sensing}},
  author = {Zhifu Tian and Tao Hu and Chaoyang Niu and Di Wu and Shu Wang},
  booktitle = {CVPR 2025},
  year = {2025},
  doi = {10.1109/CVPR52734.2025.00228},
  eprint = {2503.13241},
  archivePrefix = {arXiv}
}