New AI models tackle image and video restoration with advanced techniques
作者PulseAugur 编辑部·[9 个来源]·
Researchers have developed several new methods for image and video restoration tasks. One approach, Continuous Expert Assembly (CEA), uses a dynamic parameterization framework to adapt to diverse local degradation patterns in images. Another method integrates segmentation models like SAM2 to derive region-distinguishable priors for more accurate video frame interpolation. Additionally, a benchmark has been created for evaluating multi-frame image restoration under severe refractive warping, and a hybrid Transformer-State-Space Model framework aims to improve restoration efficiency on edge hardware.
AI
影响
Advances in image and video restoration techniques could improve visual quality in various applications, from media to surveillance.
排序理由
Multiple research papers published on arXiv detailing new methods and benchmarks for image and video restoration.
arXiv:2312.15868v3 Announce Type: replace Abstract: In existing restoration-oriented Video Frame Interpolation (VFI) approaches, the motion estimation between neighboring frames plays a crucial role. However, the estimation accuracy in existing methods remains a challenge, primar…
arXiv:2605.06127v1 Announce Type: new Abstract: Real-world image degradation is often unknown, spatially non-uniform, and compositional, requiring all-in-one restoration models to adapt a single set of weights to diverse local corruption patterns without test-time degradation lab…
Real-world image degradation is often unknown, spatially non-uniform, and compositional, requiring all-in-one restoration models to adapt a single set of weights to diverse local corruption patterns without test-time degradation labels. Existing methods typically modulate a share…
arXiv cs.CV
TIER_1English(EN)·Maxim V. Shugaev, Md Reshad Ul Hoque, Bridget Kennedy, Joseph T. Riley, Fiona Hwang, Justin Hagen, Harvir Ghuman, Ethan Garcia-O'Donnell, Syed Noor Qadri, Freddie Santiago, Mun Wai Lee·
arXiv:2605.05079v1 Announce Type: new Abstract: Video sequence capturing through refractive dynamic media, such as a turbulent air or water surface, often suffer from severe geometric distortions and temporal instability. While recent advances address mild atmospheric turbulence,…
Video sequence capturing through refractive dynamic media, such as a turbulent air or water surface, often suffer from severe geometric distortions and temporal instability. While recent advances address mild atmospheric turbulence, no existing benchmarks systematically evaluate …
arXiv:2605.02794v1 Announce Type: new Abstract: We propose a modular framework for hybrid image restoration that integrates transformer and state-space model (SSM) blocks with a focus on improving runtime efficiency on edge hardware. While transformers provide strong global model…
arXiv:2605.01236v1 Announce Type: new Abstract: Unified image restoration using a single model often faces task interference due to diverse degradations. To address this, we propose DACG-IR (Degradation-Aware Adaptive Context Gating), which enables explicit perception of degradat…
We propose a modular framework for hybrid image restoration that integrates transformer and state-space model (SSM) blocks with a focus on improving runtime efficiency on edge hardware. While transformers provide strong global modeling through self-attention, their attention kern…
arXiv cs.CV
TIER_1English(EN)·Hebaixu Wang, Jing Zhang, Haoyang Chen, Haonan Guo, Di Wang, Jiayi Ma, Bo Du·
arXiv:2510.23116v4 Announce Type: replace Abstract: Diffusion bridge models establish probabilistic paths between arbitrary paired distributions and exhibit great potential for universal image restoration. Most existing methods merely treat them as simple variants of stochastic i…
