CVPR 2026 impressions

Conferences before

 

Conferences before

 

Solution:

Workshop on bitter lessons!

• Three sessions with three speakers talking about the bitter lesson (and more broadly the AI “explosion”) + panel discussions

• Jon Barron: “3D = bitter lessonned?”

• Vincent Sitzmann: “3D (and more broadly mid-level vision) is useless”

• Bharath Hahirahan: “Mid level vision is useful”

Is 3D useful?

Two contradictory signals:

• Workshop on bitter lessons: avoid 3D when you don’t need it explicitely

• Last 3 best papers:

  • Dust3r (unofficial)
  • VGGT
  • D4RT

• Small scale experiments on transfer learning for VGG-\(\Omega\) (+ numerous papers on improving video generation using explicit 3D).

Random takes

• 3D for vision might be dead, but vision for 3D is definitely alive (Li et al. 2026; Liu, Lang, and Hanocka 2026; Xiao et al. 2026).

• “3D is dead” ≠ “Geometry is dead” (Chiang et al. 2026; Luo et al. 2026)

• 2026: LLMs still don’t improve the quality of orals or posters

• TRELLIS structured latent space enables numerous applications (tokenization(Dutt et al. 2026), editing(Hu et al. 2026), morphing(Sun et al. 2026), improving trellis(Xia et al. 2026) (again), …)

• Weak signal that DinoV3 is not that better that DinoV2

BrickNet: Graph-Backed Generative Brick Assembly

 

BrickNet: Graph-Backed Generative Brick Assembly

 

Can You Learn to See Without Images? Procedural Warm-Up for Vision Transformers

 

Can You Learn to See Without Images? Procedural Warm-Up for Vision Transformers

 

 

Chiang, Chiensheng, Kuan-Hsun Tu, Jia-Wei Liao, Cheng-Fu Chou, and Tsung-Wei Ke. 2026. “COT-FM: Cluster-Wise Optimal Transport Flow Matching.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Dutt, Niladri Shekhar, Zifan Shi, Paul Guerrero, Chun-Hao Paul Huang, Duygu Ceylan, Niloy J. Mitra, and Xuelin Chen. 2026. “LoST: Level of Semantics Tokenization for 3D Shapes.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Hu, Shimin, Yuanyi Wei, Fei Zha, Yudong Guo, and Juyong Zhang. 2026. “Easy3E: Feed-Forward 3D Asset Editing via Rectified Voxel Flow.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Li, Zizhang, Cheng Zhang, Zhengqin Li, Henry Howard-Jenkins, Zhaoyang Lv, Chen Geng, Jiajun Wu, Richard Newcombe, Jakob Engel, and Zhao Dong. 2026. “ART: Articulated Reconstruction Transformer.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Liu, Richard, Itai Lang, and Rana Hanocka. 2026. “Deep Feature Deformation Weights.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Luo, Gaoxiang, Frank Cole, Sihang Zhang, Yuxiang Wan, Yulong Lu, and Ju Sun. 2026. “Flow Matching for Multimodal Distributions.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Sun, Xiaokun, Zeyu Cai, Hao Tang, Ying Tai, Jian Yang, and Zhenyu Zhang. 2026. “MorphAny3D: Unleashing the Power of Structured Latent in 3D Morphing.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Xia, Jiatong, Zicheng Duan, Anton van den Hengel, and Lingqiao Liu. 2026. “Points-to-3D: Structure-Aware 3D Generation with Point Cloud Priors.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Xiao, Qinfeng, Guofeng Mei, Bo Yang, Liying Zhang, Jian Zhang, and Kit-lun Yick. 2026. “Universal 3D Shape Matching via Coarse-to-Fine Language Guidance.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).