Spacetools Cvpr 2026

SpaceTools (CVPR 2026)

Video presentation for "STALL: Training-free Detection of Generated Videos via Spatial-Temporal Likelihoods", presented at ...

Joonki Min, Chaeyun Kim, Hyungwook Choi, Yejin Kim, Kihyun Kim, Yohan Jo, Joonseok Lee. Fine-Grained Multi-Image Object ...

Adaptive Spatial-Temporal Window: Unlocking the Potential of Event Cameras in Heterogeneous Velocity Scenarios Zhipeng Sui, ...

Rameen Abdal, James Burgess, Sergey Tulyakov, Kuan-Chieh Wang Snap Research , Stanford University ...

Large-Scale Codec Avatars (LCA): The Unreasonable Effectiveness of Large-Scale Avatar Pretraining

Disentangle-then-Align: Non-Iterative Hybrid Multimodal Image Registration via Cross-Scale Feature Disentanglement.

MixerCSeg: An Efficient Mixer Architecture for Crack Segmentation via Decoupled Mamba Attention.

Towards Open-Vocabulary Industrial Defect Understanding with a Large-Scale Multimodal Dataset.

Hakyeong Kim, Ruicheng Wang, Chengtang Yao, Jiaolong Yang, Min H. Kim (

AnchorSplat: Feed-Forward 3D Gaussian Splatting with 3D Geometric Priors Paper: https://arxiv.org/abs/2604.07053.

CVPR 2026: Learning 3D Shape Fidelity Metric from Real-world Distortions

TokenLight is a method for image relighting that gives you precise, continuous control over lighting attributes like intensity, color, ...

Best Segmentation Buddies for Image-Shape Correspondence Itai Lang, Dongwei Lyu, Dale Decatur, Rana Hanocka University ...

CVPR 2026

[

The 5-minute introduction video of IntrinsicWeather.

We introduce our

TAPE: Task-Adaptive Prototype Evolution in Audio-Language Models for Fully Few-shot Class-incremental Audio Classification.

arxiv.org/abs/2512.14884.