Research map

3D vision: research map

Reconstruction, NeRF, Gaussian splatting, pose and depth.

6,605 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026), in 6 topics. Within all five venues, its share held steady from 6.7% in 2023–24 to 6.1% in 2025–26 (1,781 → 2,765 papers at ICML, NeurIPS, CVPR and AAAI, the venues with data for all four years).

Explore 3D vision in the interactive map

Topics

• 3D generation · 1,606 papers
  Learning 3D reconstruction from images, Semantic scene completion, 2D-to-3D pretrained representations
• Depth and stereo · 1,365 papers
  Feed-forward 3D reconstruction, Camera model and calibration for SfM, Learned optical and scene flow
• Face and shape · 1,286 papers
  Neural implicit surface reconstruction, Textured mesh generation, Learned mesh and geometric representations
• Human pose estimation · 1,051 papers
  Depth-based hand pose regression, Learned camera pose estimation, Category-level object pose estimation
• Gaussian splatting · 841 papers
  Compact and efficient Gaussian splatting, Regularization and primitive design for 3DGS, Sparse-view Gaussian splatting for novel views
• NeRF and radiance fields · 456 papers
  Sparse-input NeRF regularization, Dynamic and editable NeRF, Single-view and streaming radiance fields

Most cited papers in 3D vision

• Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields (CVPR 2017 · 7,514 citations)
• Structure-From-Motion Revisited (CVPR 2016 · 6,291 citations)
• Deep High-Resolution Representation Learning for Human Pose Estimation (CVPR 2019 · 5,835 citations)
• ScanNet: Richly-Annotated 3D Reconstructions of Indoor Scenes (CVPR 2017 · 4,275 citations)
• DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (CVPR 2019 · 3,528 citations)
• Unsupervised Monocular Depth Estimation With Left-Right Consistency (CVPR 2017 · 3,325 citations)
• Scalability in Perception for Autonomous Driving: Waymo Open Dataset (CVPR 2020 · 3,252 citations)
• Unsupervised Learning of Depth and Ego-Motion From Video (CVPR 2017 · 2,885 citations)
• SuperGlue: Learning Feature Matching With Graph Neural Networks (CVPR 2020 · 2,845 citations)
• Occupancy Networks: Learning 3D Reconstruction in Function Space (CVPR 2019 · 2,830 citations)
• End-to-End Recovery of Human Shape and Pose (CVPR 2018 · 1,958 citations)
• Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (CVPR 2022 · 1,901 citations)