Large-scale text-to-image models are attractive backbones for dense prediction because RGB generation pretraining learns rich semantic, structural, and geometric priors. Existing generative and editing approaches reuse these priors by casting dense prediction as target generation: annotations such as depth, normals, alpha mattes, masks, and heatmaps are encoded into an RGB-trained VAE latent space and decoded back as image-like targets.
arXiv:2607. 02131v1 Announce Type: cross Abstract: Restoring archival film remains a fundamentally challenging problem due to the absence of paired training data and the lack of standardized evaluation benchmarks.
By Miko{\l}aj Jastrz\k{e}bski, Dawid Glinkowski, Dawid Zieli\'nski, Daniel Borkowski, Wojciech Koz{\l}owski, Kamil Adamczewski
Camera intrinsics are vital for recovering 3D structure from 2D video. However, most 3D algorithms assume fixed intrinsics throughout a video, an assumption that often fails for real-world in-the-wild videos.
Pixel-space generative models bypass lossy latent compression, yet necessitate joint learning of global structure and fine-grained details in a high-dimensional space. Standard flow matching interpolates noise toward a fixed clean-image endpoint, leaving the spectral evolution to be learned implicitly.
arXiv:2604. 10359v3 Announce Type: replace-cross Abstract: Low-light image enhancement (LLIE) aims to restore natural visibility, color fidelity, and structural detail under severe illumination degradation.
By Alexandru Brateanu, Tingting Mu, Codruta Ancuti, Cosmin Ancuti
arXiv:2405. 01558v4 Announce Type: replace-cross Abstract: Rendering holograms for holographic displays is often an iterative and computationally costly process.
By Yicheng Zhan, Liang Shi, Wojciech Matusik, Qi Sun, Kaan Ak\c{s}it
arXiv:2604. 23094v2 Announce Type: replace-cross Abstract: Portrait relighting is a low-level vision problem in which physically plausible illumination transfer, identity preservation, and compact real-time inference must be considered together.
By Qian Huang, Mayoore Selvarasa Jaiswal, Zhen Zhong, Rochelle Pereira, Jianyuan Min
arXiv:2607. 17099v1 Announce Type: cross Abstract: Recent geometric foundation models (e.
By Feng Xue, Wu Chen, Mingshuai Zhao, Guofeng Zhong, Anlong Ming, Haozhe Wang, Dianqiao Lei, Zhaowen Lin, Haiyang Zhang, Nicu Sebe
arXiv:2604. 05182v2 Announce Type: replace-cross Abstract: We introduce the Large Sparse Reconstruction Model to study how scaling transformer context windows affects feed-forward 3D reconstruction.
By Zhengqin Li, Cheng Zhang, Jakob Engel, Zhao Dong
arXiv:2607. 08794v1 Announce Type: cross Abstract: Sand boils on earthen levees are safety-critical defects, but pixel-level detection is limited by scarce annotations.
By Padam Jung Thapa, Abdullah Bin Naeem, Ayon Dey, Anav Katwal, Md Tamjidul Hoque
Subject-driven image personalization---generating new images that preserve the identity of one or several reference subjects in novel scenes---is a foundational capability for modern visual content creation. It is currently dominated by generalized methods that fine-tune a pretrained multimodal diffusion transformer (MMDiT) on hundreds of thousands to millions of paired \emph{(reference, composed-target)} examples, where each composed target is a synthesized image of the subject in a novel scene.
arXiv:2606. 12987v1 Announce Type: cross Abstract: Action-conditioned world models let an autonomous vehicle predict future camera scenes from its own planned controls, enabling planning and simulation without real-world rollouts, but at compact, trainable scale the futures are ambiguous and the field's standard distortion metrics actively mislead: they reward a blurry regression mean over a realistic prediction.
By Ruslan Sharifullin, Benjamin Jiang, Kai Xi Chew