Depth from Defocus via Direct Optimization

• URL: http://arxiv.org/abs/2602.18509v2
• Date: Thu, 26 Feb 2026 05:47:11 GMT
• Title: Depth from Defocus via Direct Optimization
• Authors: Holly Jackson, Caleb Adams, Ignacio Lopez-Francos, Benjamin Recht,
• Abstract summary: We show that with contemporary optimization methods and reasonable computing resources, a global approach to depth from defocus is feasible.<n>We show that alternating between convex optimization and parallel grid search can effectively solve the depth-from-defocus problem at higher resolutions than current deep learning methods.
• Score: 4.661494043076582
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Though there exists a reasonable forward model for blur based on optical physics, recovering depth from a collection of defocused images remains a computationally challenging optimization problem. In this paper, we show that with contemporary optimization methods and reasonable computing resources, a global optimization approach to depth from defocus is feasible. Our approach rests on alternating minimization. When holding the depth map fixed, the forward model is linear with respect to the all-in-focus image. When holding the all-in-focus image fixed, the depth at each pixel can be computed independently, enabling embarrassingly parallel computation. We show that alternating between convex optimization and parallel grid search can effectively solve the depth-from-defocus problem at higher resolutions than current deep learning methods. We demonstrate our approach on benchmark datasets with synthetic and real defocus blur and show promising results compared to prior approaches. Our code is available at github.com/hollyjackson/dfd.

Related papers

• Depth-Supervised Fusion Network for Seamless-Free Image Stitching [17.758594808190264]
  Image stitching synthesizes images captured from multiple perspectives into a single image with a broader field of view.<n>The significant variations in object depth often lead to large parallax, resulting in ghosting and misalignment in the stitched results.<n>To address this, we propose a depth-consistency-constrained seamless-free image stitching method.
  arXiv  Detail & Related papers  (2025-10-24T12:36:08Z)
• Pixel-Aligned Multi-View Generation with Depth Guided Decoder [86.1813201212539]
  We propose a novel method for pixel-level image-to-multi-view generation. Unlike prior work, we incorporate attention layers across multi-view images in the VAE decoder of a latent video diffusion model. Our model enables better pixel alignment across multi-view images.
  arXiv  Detail & Related papers  (2024-08-26T04:56:41Z)
• Towards Real-World Focus Stacking with Deep Learning [97.34754533628322]
  We introduce a new dataset consisting of 94 high-resolution bursts of raw images with focus bracketing. This dataset is used to train the first deep learning algorithm for focus stacking capable of handling bursts of sufficient length for real-world applications.
  arXiv  Detail & Related papers  (2023-11-29T17:49:33Z)
• Single Image Depth Prediction Made Better: A Multivariate Gaussian Take [163.14849753700682]
  We introduce an approach that performs continuous modeling of per-pixel depth. Our method's accuracy (named MG) is among the top on the KITTI depth-prediction benchmark leaderboard.
  arXiv  Detail & Related papers  (2023-03-31T16:01:03Z)
• Single image deep defocus estimation and its applications [82.93345261434943]
  We train a deep neural network to classify image patches into one of the 20 levels of blurriness. The trained model is used to determine the patch blurriness which is then refined by applying an iterative weighted guided filter. The result is a defocus map that carries the information of the degree of blurriness for each pixel.
  arXiv  Detail & Related papers  (2021-07-30T06:18:16Z)
• Efficient and Differentiable Shadow Computation for Inverse Problems [64.70468076488419]
  Differentiable geometric computation has received increasing interest for image-based inverse problems. We propose an efficient yet efficient approach for differentiable visibility and soft shadow computation. As our formulation is differentiable, it can be used to solve inverse problems such as texture, illumination, rigid pose, and deformation recovery from images.
  arXiv  Detail & Related papers  (2021-04-01T09:29:05Z)
• Real-Time, Deep Synthetic Aperture Sonar (SAS) Autofocus [34.77467193499518]
  Synthetic aperture sonar (SAS) requires precise time-of-flight measurements of the transmitted/received waveform to produce well-focused imagery. To overcome this, an emphautofocus algorithm is employed as a post-processing step after image reconstruction to improve image focus. We propose a deep learning technique to overcome these limitations and implicitly learn the weighting function in a data-driven manner.
  arXiv  Detail & Related papers  (2021-03-18T15:16:29Z)
• Towards Reducing Severe Defocus Spread Effects for Multi-Focus Image Fusion via an Optimization Based Strategy [22.29205225281694]
  Multi-focus image fusion (MFF) is a popular technique to generate an all-in-focus image. This paper presents an optimization-based approach to reduce defocus spread effects. Experiments conducted on the real-world dataset verify superiority of the proposed model.
  arXiv  Detail & Related papers  (2020-12-29T09:26:41Z)
• Deep Autofocus for Synthetic Aperture Sonar [28.306713374371814]
  In this letter, we demonstrate the potential of machine learning, specifically deep learning, to address the autofocus problem. We formulate the problem as a self-supervised, phase error estimation task using a deep network we call Deep Autofocus. Our results demonstrate Deep Autofocus can produce imagery that is perceptually as good as benchmark iterative techniques but at a substantially lower computational cost.
  arXiv  Detail & Related papers  (2020-10-29T15:31:15Z)
• A Flexible Framework for Designing Trainable Priors with Adaptive Smoothing and Game Encoding [57.1077544780653]
  We introduce a general framework for designing and training neural network layers whose forward passes can be interpreted as solving non-smooth convex optimization problems. We focus on convex games, solved by local agents represented by the nodes of a graph and interacting through regularization functions. This approach is appealing for solving imaging problems, as it allows the use of classical image priors within deep models that are trainable end to end.
  arXiv  Detail & Related papers  (2020-06-26T08:34:54Z)
• AcED: Accurate and Edge-consistent Monocular Depth Estimation [0.0]
  Single image depth estimation is a challenging problem. We formulate a fully differentiable ordinal regression and train the network in end-to-end fashion. A novel per-pixel confidence map computation for depth refinement is also proposed.
  arXiv  Detail & Related papers  (2020-06-16T15:21:00Z)

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.