mlsanityData to 7 Oct 2026

A General and Adaptive Robust Loss Function

arXiv 1701.03077 ↗First posted 11 Jan 2017cs.CV cs.LG stat.MLcited by 236

Not on the map. Why?

Abstract (arXiv)

We present a generalization of the Cauchy/Lorentzian, Geman-McClure, Welsch/Leclerc, generalized Charbonnier, Charbonnier/pseudo-Huber/L1-L2, and L2 loss functions. By introducing robustness as a continuous parameter, our loss function allows algorithms built around robust loss minimization to be generalized, which improves performance on basic vision tasks such as registration and clustering. Interpreting our loss as the negative log of a univariate density yields a general probability distribution that includes normal and Cauchy distributions as special cases. This probabilistic interpretation enables the training of neural networks in which the robustness of the loss automatically adapts itself during training, which improves performance on learning-based tasks such as generative image synthesis and unsupervised monocular depth estimation, without requiring any manual parameter tuning.

Key facts. Cited by 236 papers in the map (data through 7 Oct 2026). First posted on arXiv 11 Jan 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 39; 2022: 40; 2023: 41; 2024: 31; 2025: 41; 2026: 44.

236 papers.

Sort:

By area:

  1. 2023Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fieldscited by 667
  2. 2021Baking Neural Radiance Fields for Real-Time View Synthesiscited by 465
  3. 2023RoMa: Robust Dense Feature Matchingcited by 296
  4. 2025MapAnything: Universal Feed-Forward Metric 3D Reconstructioncited by 295
  5. 2023BakedSDF: Meshing Neural SDFs for Real-Time View Synthesiscited by 220
  6. 2021Human Pose Regression with Residual Log-likelihood Estimationcited by 131
  7. 2023FoundPose: Unseen Object Pose Estimation with Foundation Featurescited by 109
  8. 2023RobustNeRF: Ignoring Distractors with Robust Lossescited by 101
  9. 2023EMDB: The Electromagnetic Database of Global 3D Human Pose and Shape in the Wildcited by 95
  10. 2022Theseus: A Library for Differentiable Nonlinear Optimizationcited by 64
  11. 2025HaWoR: World-Space Hand Motion Reconstruction from Egocentric Videoscited by 62
  12. 2021OmniDet: Surround View Cameras based Multi-task Visual Perception Network for Autonomous Drivingcited by 57
38 more
  1. 2022NeRF-Pose: A First-Reconstruct-Then-Regress Approach for Weakly-supervised 6D Object Pose Estimationcited by 57
  2. 2022Scaleformer: Iterative Multi-scale Refining Transformers for Time Series Forecastingcited by 54
  3. 2022SparsePose: Sparse-View Camera Pose Regression and Refinementcited by 53
  4. 2021Meta-learning PINN loss functionscited by 51
  5. 2024Robot See Robot Do: Imitating Articulated Object Manipulation with Monocular 4D Reconstructioncited by 51
  6. 2025RoMa v2: Harder Better Faster Denser Feature Matchingcited by 44
  7. 2023In Search of Lost Online Test-time Adaptation: A Surveycited by 43
  8. 2022High Probability Bounds for a Class of Nonconvex Algorithms with AdaGrad Stepsizecited by 40
  9. 2023FLARE: Fast Learning of Animatable and Relightable Mesh Avatarscited by 39
  10. 2022Adapting to Mixing Time in Stochastic Optimization with Markovian Datacited by 35
  11. 2025UFM: A Simple Path towards Unified Dense Correspondence with Flowcited by 35
  12. 2022EM-driven unsupervised learning for efficient motion segmentationcited by 33
  13. 2022Surround-view Fisheye Camera Perception for Automated Driving: Overview, Survey and Challengescited by 33
  14. 2023Object Pose Estimation with Statistical Guarantees: Conformal Keypoint Detection and Geometric Uncertainty Propagationcited by 33
  15. 2021Generalized Huber Loss for Robust Learning and its Efficient Minimization for a Robust Statisticscited by 32
  16. 2022DPODv2: Dense Correspondence-Based 6 DoF Pose Estimationcited by 32
  17. 2021Scene Synthesis via Uncertainty-Driven Attribute Synchronizationcited by 29
  18. 2023Explicitly Minimizing the Blur Error of Variational Autoencoderscited by 29
  19. 2021MULLS: Versatile LiDAR SLAM via Multi-metric Linear Least Squarecited by 27
  20. 2021SVDistNet: Self-Supervised Near-Field Distance Estimation on Surround View Fisheye Camerascited by 25
  21. 2021Defocus Map Estimation and Deblurring from a Single Dual-Pixel Imagecited by 25
  22. 2024Robust Reinforcement Learning from Corrupted Human Feedbackcited by 25
  23. 2021A Comprehensive Review of Deep Learning-based Single Image Super-resolutioncited by 23
  24. 2021NTIRE 2021 Depth Guided Image Relighting Challengecited by 23
  25. 2022Few-shot Learning with Noisy Labelscited by 23
  26. 2023NEF: Neural Edge Fields for 3D Parametric Curve Reconstruction from Multi-view Imagescited by 22
  27. 2024LiteVAE: Lightweight and Efficient Variational Autoencoders for Latent Diffusion Modelscited by 22
  28. 2024SuperGaussian: Repurposing Video Models for 3D Super Resolutioncited by 22
  29. 2021Certifiably Optimal Outlier-Robust Geometric Perception: Semidefinite Relaxations and Scalable Global Optimizationcited by 21
  30. 2023DReg-NeRF: Deep Registration for Neural Radiance Fieldscited by 21
  31. 20243Doodle: Compact Abstraction of Objects with 3D Strokescited by 21
  32. 2022Spach Transformer: Spatial and Channel-wise Transformer Based on Local and Global Self-attentions for PET Image Denoisingcited by 19
  33. 2025Decoupled MeanFlow: Turning Flow Models into Flow Maps for Accelerated Samplingcited by 19
  34. 2021Squareplus: A Softplus-Like Algebraic Rectifiercited by 18
  35. 2021Neural Reprojection Error: Merging Feature Learning and Camera Pose Estimationcited by 17
  36. 2023HMD-NeMo: Online 3D Avatar Motion Generation From Sparse Observationscited by 17
  37. 2021Learning to Fit Morphable Modelscited by 16
  38. 2025R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localizationcited by 16

50 most cited of 236 All 236 (JSON)