How to cheat with metrics in single-image HDR reconstruction

• URL: http://arxiv.org/abs/2108.08713v1
• Date: Thu, 19 Aug 2021 14:29:15 GMT
• Title: How to cheat with metrics in single-image HDR reconstruction
• Authors: Gabriel Eilertsen, Saghi Hajisharif, Param Hanji, Apostolia Tsirikoglou, Rafal K. Mantiuk, Jonas Unger
• Abstract summary: Single-image high dynamic range (SI-The) reconstruction has recently emerged as a problem well-suited for deep learning methods. This paper highlights that such improvements in objective metrics do not necessarily translate to visually superior images. We show how such results are not representative of the perceived quality and that SI- reconstruction needs better evaluation protocols.
• Score: 19.361293438802264
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Single-image high dynamic range (SI-HDR) reconstruction has recently emerged as a problem well-suited for deep learning methods. Each successive technique demonstrates an improvement over existing methods by reporting higher image quality scores. This paper, however, highlights that such improvements in objective metrics do not necessarily translate to visually superior images. The first problem is the use of disparate evaluation conditions in terms of data and metric parameters, calling for a standardized protocol to make it possible to compare between papers. The second problem, which forms the main focus of this paper, is the inherent difficulty in evaluating SI-HDR reconstructions since certain aspects of the reconstruction problem dominate objective differences, thereby introducing a bias. Here, we reproduce a typical evaluation using existing as well as simulated SI-HDR methods to demonstrate how different aspects of the problem affect objective quality metrics. Surprisingly, we found that methods that do not even reconstruct HDR information can compete with state-of-the-art deep learning methods. We show how such results are not representative of the perceived quality and that SI-HDR reconstruction needs better evaluation protocols.

Related papers

• A Cycle Ride to HDR: Semantics Aware Self-Supervised Framework for Unpaired LDR-to-HDR Image Translation [0.0]
  Low Dynamic Range (LDR) to High Dynamic Range () image translation is an important computer vision problem. Most current state-of-the-art methods require high-quality paired LDR, datasets for model training. We propose a modified cycle-consistent adversarial architecture and utilize unpaired LDR, datasets for training.
  arXiv  Detail & Related papers  (2024-10-19T11:11:58Z)
• Single-Image HDR Reconstruction by Multi-Exposure Generation [8.656080193351581]
  We propose a weakly supervised learning method that inverts the physical image formation process for HDR reconstruction. Our neural network can invert the camera response to reconstruct pixel irradiance before synthesizing multiple exposures. Our experiments show that our proposed model can effectively reconstruct HDR images.
  arXiv  Detail & Related papers  (2022-10-28T05:12:56Z)
• Hierarchical Similarity Learning for Aliasing Suppression Image Super-Resolution [64.15915577164894]
  A hierarchical image super-resolution network (HSRNet) is proposed to suppress the influence of aliasing. HSRNet achieves better quantitative and visual performance than other works, and remits the aliasing more effectively.
  arXiv  Detail & Related papers  (2022-06-07T14:55:32Z)
• Image Reconstruction from Events. Why learn it? [11.773972029187433]
  We show how tackling the joint problem of motion estimation leads us to model event-based image reconstruction as a linear inverse problem. We propose classical and learning-based image priors can be used to solve the problem and remove artifacts from the reconstructed images.
  arXiv  Detail & Related papers  (2021-12-12T14:01:09Z)
• Image Quality Assessment using Contrastive Learning [50.265638572116984]
  We train a deep Convolutional Neural Network (CNN) using a contrastive pairwise objective to solve the auxiliary problem. We show through extensive experiments that CONTRIQUE achieves competitive performance when compared to state-of-the-art NR image quality models. Our results suggest that powerful quality representations with perceptual relevance can be obtained without requiring large labeled subjective image quality datasets.
  arXiv  Detail & Related papers  (2021-10-25T21:01:00Z)
• Learning Conditional Knowledge Distillation for Degraded-Reference Image Quality Assessment [157.1292674649519]
  We propose a practical solution named degraded-reference IQA (DR-IQA) DR-IQA exploits the inputs of IR models, degraded images, as references. Our results can even be close to the performance of full-reference settings.
  arXiv  Detail & Related papers  (2021-08-18T02:35:08Z)
• Joint reconstruction and bias field correction for undersampled MR imaging [7.409376558513677]
  Undersampling the k-space in MRI allows saving precious acquisition time, yet results in an ill-posed inversion problem. Deep learning schemes are susceptible to differences between the training data and the image to be reconstructed at test time. In this work, we address the sensitivity of the reconstruction problem to the bias field and propose to model it explicitly in the reconstruction.
  arXiv  Detail & Related papers  (2020-07-26T12:58:34Z)
• Single-Image HDR Reconstruction by Learning to Reverse the Camera Pipeline [100.5353614588565]
  We propose to incorporate the domain knowledge of the LDR image formation pipeline into our model. We model the HDRto-LDR image formation pipeline as the (1) dynamic range clipping, (2) non-linear mapping from a camera response function, and (3) quantization. We demonstrate that the proposed method performs favorably against state-of-the-art single-image HDR reconstruction algorithms.
  arXiv  Detail & Related papers  (2020-04-02T17:59:04Z)
• Deep Face Super-Resolution with Iterative Collaboration between Attentive Recovery and Landmark Estimation [92.86123832948809]
  We propose a deep face super-resolution (FSR) method with iterative collaboration between two recurrent networks. In each recurrent step, the recovery branch utilizes the prior knowledge of landmarks to yield higher-quality images. A new attentive fusion module is designed to strengthen the guidance of landmark maps.
  arXiv  Detail & Related papers  (2020-03-29T16:04:48Z)
• Characteristic Regularisation for Super-Resolving Face Images [81.84939112201377]
  Existing facial image super-resolution (SR) methods focus mostly on improving artificially down-sampled low-resolution (LR) imagery. Previous unsupervised domain adaptation (UDA) methods address this issue by training a model using unpaired genuine LR and HR data. This renders the model overstretched with two tasks: consistifying the visual characteristics and enhancing the image resolution. We formulate a method that joins the advantages of conventional SR and UDA models.
  arXiv  Detail & Related papers  (2019-12-30T16:27:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.