Lune

CVPR2026Top-tier venue

VGGT-Ω

Jianyuan Wang, Minghao Chen, Shangzhan Zhang, Nikita Karaev, Johannes Schönberger, Patrick Labatut, Piotr Bojanowski, David Novotný, Andrea Vedaldi, Christian Rupprecht

Abstract

Recent feed-forward reconstruction models, such as VGGT, have proven competitive with traditional optimizationbased reconstructors while also providing geometry-aware features useful for other tasks. Here, we show that the quality of these models scales predictably with model and data size. We do so by introducing VGGT-Ω, which substantially improves reconstruction accuracy, efficiency, and capabilities for both static and dynamic scenes. To enable training this model at an unprecedented scale, we introduce architectural changes that improve training efficiency, a high-quality data annotation pipeline that supports dynamic scenes, and a self-supervised learning protocol. We simplify VGGT's architecture by using a single dense prediction head with multi-task supervision and removing the expensive high-resolution convolutional layers. We also use registers to aggregate scene information into a compact representation and introduce register attention, which restricts inter-frame information exchange to these registers, in part replacing global attention. In this way, during training, VGGT-Ω uses only ∼30% of the GPU memory of its predecessor, which allows us to train VGGT-Ω with 15× more supervised data than prior work and to leverage vast amounts of unlabeled video data. VGGT-Ω achieves strong results for reconstruction of static and dynamic scenes across multiple benchmarks, e.g., improving over the previous best camera estimation accuracy on Sintel by 77%. We also show that the learned registers can improve vision-language-action models and support alignment with language, suggesting that reconstruction can be a powerful and scalable proxy task for spatial understanding.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Sigmoid Loss for Language Image Pre-Training cites Learning Transferable Visual Models From Natural Language Supervision
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on98

  1. Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
  2. Emerging Properties in Self-Supervised Vision TransformersMathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou et al.ICCV 2021 · 8,921 citations
  3. Sigmoid Loss for Language Image Pre-TrainingXiaohua Zhai, Basil Mustafa, Alexander Kolesnikov, Lucas BeyerICCV 2023 · 2,932 citations
  4. Vision Transformers for Dense PredictionRené Ranftl, Alexey Bochkovskiy, Vladlen KoltunICCV 2021 · 2,647 citations
  5. Habitat: A Platform for Embodied AI ResearchManolis Savva, Jitendra Malik, Devi Parikh, Dhruv Batra et al.ICCV 2019 · 1,863 citations

Related papers

  1. VGGT: Visual Geometry Grounded TransformerJianyuan Wang, Minghao Chen, Nikita Karaev, Andrea Vedaldi et al.CVPR 2025
  2. FlashVGGT: Efficient and Scalable Visual Geometry Transformers with Compressed Descriptor AttentionZipeng Wang, Dan XuCVPR 2026 · 14 citations
  3. Selfi: Self-improving Reconstruction Engine via 3D Geometric Feature AlignmentYouming Deng, Songyou Peng, Junyi Zhang, Kathryn Heal et al.CVPR 2026 · 4 citations
  4. QVGGT: Post-Training Quantized Visual Geometry Grounded TransformerZhizhen Pan, Hesong Wang, Huan WangCVPR 2026 · 2 citations
  5. Scal3R: Scalable Test-Time Training for Large-Scale 3D ReconstructionTao Xie, Peishan Yang, Yudong Jin, Yingfeng Cai et al.CVPR 2026 · 26 citations
Dusk over the sea between two cliffs drawn in fine vertical lines