mlsanityData to 14 Oct 2026

Loss Functions for Neural Networks for Image Processing

arXiv 1511.08861 ↗First posted 28 Nov 2015cs.CVcited by 41

Not on the map. Why?

Abstract (arXiv)

Neural networks are becoming central in several areas of computer vision and image processing and different architectures have been proposed to solve specific problems. The impact of the loss layer of neural networks, however, has not received much attention in the context of image processing: the default and virtually only choice is L2. In this paper, we bring attention to alternative choices for image restoration. In particular, we show the importance of perceptually-motivated losses when the resulting image is to be evaluated by a human observer. We compare the performance of several losses, and propose a novel, differentiable error function. We show that the quality of the results improves significantly with better loss functions, even when the network architecture is left unchanged.

Key facts. Cited by 41 papers in the map (data through 14 Oct 2026). First posted on arXiv 28 Nov 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2019: 1; 2020: 0; 2021: 14; 2022: 9; 2023: 4; 2024: 8; 2025: 3; 2026: 2.

41 papers.

Sort:

By area:

  1. 2021Deep learning-based synthetic-CT generation in radiotherapy and PET: a reviewcited by 28
  2. 2021SurfaceNet: Adversarial SVBRDF Estimation from a Single Imagecited by 25
  3. 2021Anomaly Detection via Self-organizing Mapcited by 18
  4. 2022Facial Action Unit Detection and Intensity Estimation from Self-supervised Representationcited by 16
  5. 2021Trusted Artificial Intelligence: Towards Certification of Machine Learning Applicationscited by 13
  6. 2022Semi-Supervised Learning with Mutual Distillation for Monocular Depth Estimationcited by 10
  7. 2021Functional Neural Networks for Parametric Image Restoration Problemscited by 9
  8. 2019Model-based Behavioral Cloning with Future Image Similarity Learningcited by 8
  9. 2021Selective Pseudo-label Clusteringcited by 8
  10. 2022Computing Multiple Image Reconstructions with a Single Hypernetworkcited by 8
  11. 2024Generating Counterfactual Trajectories with Latent Diffusion Models for Concept Discoverycited by 8
  12. 2021On the Sins of Image Synthesis Loss for Self-supervised Depth Estimationcited by 7
29 more
  1. 2024Multi-scale Restoration of Missing Data in Optical Time-series Images with Masked Spatial-Temporal Attention Networkcited by 7
  2. 2023Vision Learners Meet Web Image-Text Pairscited by 6
  3. 2024Imaging foundation model for universal enhancement of non-ideal measurement CTcited by 6
  4. 2025UltraFast-LiNET: Light-weight multi-scale shift convolutional network for real-time low-light image enhancementcited by 5
  5. 2021Ada-VSR: Adaptive Video Super-Resolution with Meta-Learningcited by 3
  6. 2021FAMINet: Learning Real-time Semi-supervised Video Object Segmentation with Steepest Optimized Optical Flowcited by 3
  7. 2025HiFi-Mamba: Dual-Stream W-Laplacian Enhanced Mamba for High-Fidelity MRI Reconstructioncited by 3
  8. 2023An Unsupervised Framework for Joint MRI Super Resolution and Gibbs Artifact Removalcited by 2
  9. 2023Hard-normal Example-aware Template Mutual Matching for Industrial Anomaly Detectioncited by 2
  10. 2024MCU-Net: A Multi-prior Collaborative Deep Unfolding Network with Gates-controlled Spatial Attention for Accelerated MR Image Reconstructioncited by 2
  11. 2026SPACE-CLIP: Spatial Perception via Adaptive CLIP Embeddings for Monocular Depth Estimationcited by 2
  12. 2021Multitask Learning for VVC Quality Enhancement and Super-Resolutioncited by 1
  13. 2021Robotic Inspection of Underground Utilities for Construction Survey Using a Ground Penetrating Radarcited by 1
  14. 2022Image quality measurements and denoising using Fourier Ring Correlationscited by 1
  15. 2022RSINet: Inpainting Remotely Sensed Images Using Triple GAN Frameworkcited by 1
  16. 2022Attention-based Knowledge Distillation in Multi-attention Tasks: The Impact of a DCT-driven Losscited by 1
  17. 2022Deep Learning Techniques for Visual Countingcited by 1
  18. 2022VoloGAN: Adversarial Domain Adaptation for Synthetic Depth Datacited by 1
  19. 2022Audio Denoising for Robust Audio Fingerprintingcited by 1
  20. 2024Geometric Constraints in Deep Learning Frameworks: A Surveycited by 1
  21. 2025HiFi-MambaV2: Hierarchical Shared-Routed MoE for High-Fidelity MRI Reconstructioncited by 1
  22. 2021Deep Learning strategies for ProtoDUNE raw data denoisingcited by 0
  23. 2021Physics Driven Deep Retinex Fusion for Adaptive Infrared and Visible Image Fusioncited by 0
  24. 2021Attention based Broadly Self-guided Network for Low light Image Enhancementcited by 0
  25. 2023Self-Supervised Versus Supervised Training for Segmentation of Organoid Imagescited by 0
  26. 2024ReWiTe: Realistic Wide-angle and Telephoto Dual Camera Fusion Dataset via Beam Splitter Camera Rigcited by 0
  27. 2024ATAC-Net: Zoomed view works better for Anomaly Detectioncited by 0
  28. 2024LightPure: Realtime Adversarial Image Purification for Mobile Devices Using Diffusion Modelscited by 0
  29. 2026Alternate loss functions and regression models that achieve robustness to outliers by modulating the learning ratecited by 0