← Back to all news
arXiv Computer Vision By Laura C. Diaz-Delgado, Emmanuel Martinez, Henry Arguello

Frozen CLIP Priors for Robust Self-Supervised Poisson Inverse Problems

Read the original on arXiv Computer Vision →

The paper introduces a self‑supervised approach for Poisson inverse imaging problems that leverages frozen CLIP RN50 features as a lightweight prior within an ADMM‑inspired unrolled solver. By decoupling data‑consistency from the prior and using a parameter‑efficient decoder, the method adapts foundation vision representations without extensive fine‑tuning. Experiments on Poisson CFA demosaicing and deblurring demonstrate competitive image quality, enhanced robustness to dataset and acquisition shifts, and self‑supervised performance close to that of supervised training.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Related stories

arXiv Computer Vision
Sep 3

SlowFast-SCI: Slow-Fast Deep Unfolding Learning for Spectral Compressive Imaging

SlowFast‑SCI introduces a dual‑speed deep‑unfolding framework for spectral compressive imaging that combines a slow, pre‑trained backbone with a fast, test‑time adaptation stage. The slow phase distills a priors‑based model into a compact fast‑unfolding network, while the fast phase embeds lightweight modules that self‑supervise at test time without retraining the backbone. This design yields significant reductions in parameters and FLOPs, improves out‑of‑distribution PSNR by up to 5.79 dB, and accelerates adaptation four‑fold, all while remaining modular enough to integrate with any existing deep‑unfolding system.

By Haijin Zeng, Xuan Lu, Jiezhang Cao, Kai Zhang, Yurong Zhang, Qiangqiang Shen, Guoqing Chao, Li Jiang, Yongyong Chen, Jingyong Su, Jie Liu
More like this →
arXiv Computer Vision
Oct 5

One Photon, Many Worlds: Posteriors and Predictions with Single-Photon Cameras

Single-photon avalanche diode (SPAD) cameras produce binary images where each pixel records zero or one photon arrival, making scene reconstruction a fundamentally one-to-many inverse problem. As more binary frames are accumulated, the uncertainty diminishes, transitioning from stochastic to near-deterministic scene understanding. The study introduces a conditional generative framework using a Hypergeometric frame‑thinning process to model this ambiguity and evaluates its impact on tasks such as character recognition, QR code decoding, and facial analysis.

By Haejoon Lee (Carnegie Mellon University), Mohit Gupta (University of Wisconsin-Madison), Vijayakumar Bhagavatula (Carnegie Mellon University), Aswin C. Sankaranarayanan (Carnegie Mellon University)
More like this →