← Back to the metrics index

Shape reconstruction / auxiliary metric · shape

Chamfer Distance

Bidirectional Chamfer Distance

Uses bidirectional nearest-point matching between predicted and ground-truth shape point clouds to measure reconstruction similarity.

Primarily evaluates NOCS, shape completion, or category-level shape reconstruction rather than 6DoF pose alone; read it alongside pose metrics.

Principle

What error does this metric penalize?

The formulas define the computation. The diagrams show whether the error concerns a 3D model, visible surfaces, the image plane, or aggregation across thresholds.

Compute nearest-neighbor distances from prediction to ground truth and from ground truth to prediction, ensuring both sides are covered.

Can reveal missing regions or local displacements, but does not guarantee pointwise topological correspondence.

DCD(N,M)=∑x∈Nmin⁡y∈M∥x−y∥22+∑y∈Mmin⁡x∈N∥x−y∥22D_{CD}(\mathcal N,\mathcal M)=\sum_{x\in\mathcal N}\min_{y\in\mathcal M}\lVert x-y\rVert_2^2+\sum_{y\in\mathcal M}\min_{x\in\mathcal N}\lVert x-y\rVert_2^2
Works may use squared or unsquared distances, normalization by point count, or different sampling densities; specify the implementation.
Predicted pointsGround-truth pointsBidirectional nearest-neighbor distances
Illustration created for this site to explain the metric's error space; it does not show measured data or a figure from the original paper.

How to compute

Computation steps

  1. 01

    Place both shapes in the same NOCS or world coordinate frame, and sample according to the protocol.

  2. 02

    Compute nearest-neighbor distances in both directions.

  3. 03

    Sum or average to obtain CD.

Strength

What it is useful for

  • Requires no exact point correspondences and suits generated or completed shapes.
  • Penalizes both missing and extraneous regions.

Caveat

Cautions when reading results

  • Low CD does not guarantee correct pose; report shape and pose separately.
  • Highly sensitive to point density, outliers, and units.

Paper atlas

Which papers report this metric?

Links are derived by normalizing the metric names used in each paper. Directly reported metrics are distinguished from metrics associated through a broader protocol.

Filter the paper library
24Related papers
24Directly reported
0Linked through a protocol
2026 Directly reported

Active Tactile Exploration for Rigid Body Pose and Shape Estimation

ICRA · Ethan K. Gordon, Bruke Baraki, Hien Bui et al.

Original terminology: Chamfer Distance / bidirectional Chamfer distance in mm
2026 Directly reported

CanoVerse: 3D Object Scalable Canonicalization and Dataset for Generation and Pose

ECCV · Li Jin, Yuchen Yang, Weikai Chen et al.

Original terminology: Chamfer Distance
2026 Directly reported

CAOA: Completion-Assisted Object-CAD Alignment

3DV · Hiranya Garbha Kumar, Minhas Kamal, Balakrishnan Prabhakaran

Original terminology: Chamfer Distance
2026 Directly reported

Follow My Hold: Hand-Object Interaction Reconstruction through Geometric Guidance

3DV · Ayce Idil Aytekin, Helge Rhodin, Rishabh Dabral et al.

Original terminology: Chamfer Distance
2026 Directly reported

GaussianArt: Unified Modeling of Geometry and Motion for Articulated Objects

3DV · Licheng Shen, Saining Zhang, Honghan Li et al.

Original terminology: Chamfer Distance
2026 Directly reported

Reconstructing Hand-Held Objects in 3D from Images and Videos

3DV · Jane Wu, Georgios Pavlakos, Georgia Gkioxari et al.

Original terminology: Chamfer Distance
2026 Directly reported

HIPPo: Harnessing Image-to-3D Priors for Model-Free Zero-Shot 6D Pose Estimation

ICRA / IEEE RA-L · Yibo Liu, Zhaodong Jiang, Binbin Xu et al.

Original terminology: Chamfer Distance / Chamfer distance
2026 Directly reported

iTACO: Interactable Digital Twins of Articulated Objects from Casually Captured RGBD Videos

3DV · Weikun Peng, Jun Lv, Cewu Lu et al.

Original terminology: Chamfer Distance
2026 Directly reported

Deep Learning-Based Object Pose Estimation: A Comprehensive Survey

IJCV · Jian Liu, Wei Sun, Hui Yang et al.

Original terminology: Chamfer Distance
2026 Directly reported

Point2Pose: Occlusion-Recovering 6D Pose Tracking and 3D Reconstruction for Multiple Unknown Objects Via 2D Point Trackers

ECCV · Tzu-Yuan Lin, Ho Jae Lee, Kevin Doherty et al.

Original terminology: Chamfer Distance
2025 Directly reported

Betsu-Betsu: Multi-View Separable 3D Reconstruction of Two Interacting Objects

3DV · Suhas Gopal, Rishabh Dabral, Vladislav Golyanik et al.

Original terminology: Chamfer Distance
2025 Directly reported

CRISP: Object Pose and Shape Estimation with Test-Time Adaptation

CVPR · Jingnan Shi, Rajat Talak, Harry Zhang et al.

Original terminology: Chamfer distance
2025 Directly reported

iFusion: Inverting Diffusion for Pose-Free Reconstruction from Sparse Views

3DV · Chin-Hsuan Wu, Yen-Chun Chen, Bolivar Solarte et al.

Original terminology: Chamfer Distance
2025 Directly reported

InterTrack: Tracking Human Object Interaction without Object Templates

3DV · Xianghui Xie, Jan Eric Lenssen, Gerard Pons-Moll

Original terminology: Chamfer Distance
2025 Directly reported

OmniShape: Zero-Shot Multi-Hypothesis Shape and Pose Estimation in the Real World

ICRA · Katherine Liu, Sergey Zakharov, Dian Chen et al.

Original terminology: Chamfer Distance
2025 Directly reported

Category-Level 6D Object Pose Estimation in Agricultural Settings Using a Lattice-Deformation Framework and Diffusion-Augmented Synthetic Data

IROS · Marios Glytsos, Panagiotis P. Filntisis, George Retsinas et al.

Original terminology: Chamfer distance
2025 Directly reported

SDFit: 3D Object Pose and Shape by Fitting a Morphable SDF to a Single Image

ICCV · Dimitrije Antić, Georgios Paschalidis, Shashank Tripathi et al.

Original terminology: 3D Chamfer distance
2025 Directly reported

SMORE: Simultaneous Map and Object REconstruction

3DV · Nathaniel Chodosh, Anish Madan, Simon Lucey et al.

Original terminology: Chamfer Distance
2024 Directly reported

Category-level Object Detection, Pose Estimation and Reconstruction from Stereo Images

ECCV · Chuanrui Zhang, Yonggen Ling, Minglei Lu et al.

Original terminology: Chamfer distance
2024 Directly reported

Neural Implicit Representation for Building Digital Twins of Unknown Articulated Objects

CVPR · Yijia Weng, Bowen Wen, Jonathan Tremblay et al.

Original terminology: Chamfer Distance / Chamfer Distance for static / moving / whole shape
2024 Directly reported

HOC-Search: Efficient CAD Model and Pose Retrieval from RGB-D Scans

3DV · Stefan Ainetter, Sinisa Stekovic, Friedrich Fraundorfer et al.

Original terminology: Chamfer Distance
2024 Directly reported

L4D-Track: Language-to-4D Modeling Towards 6-DoF Tracking and Shape Reconstruction in 3D Point Cloud Stream

CVPR · Jingtao Sun, Yaonan Wang, Mingtao Feng et al.

Original terminology: Chamfer Distance
2024 Directly reported

Living Scenes: Multi-object Relocalization and Reconstruction in Changing 3D Environments

CVPR · Liyuan Zhu, Shengyu Huang, Konrad Schindler et al.

Original terminology: Chamfer distance
2024 Directly reported

NeuSurfEmb: A Complete Pipeline for Dense Correspondence-Based 6D Object Pose Estimation without CAD Models

IROS · Francesco Milano, Jen Jen Chung, Hermann Blum et al.

Original terminology: forward Chamfer distance
Sources for definitions and protocolsExpand to view original papers, surveys, and official evaluation pages.