# Robust Shape from Focus via Multiscale Directional Dilated Laplacian and Recurrent Network

arXiv: https://arxiv.org/abs/2512.10498 · HTML: https://mlsanity.com/paper/2512.10498/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 59 references that are in the map.
- First posted on arXiv 11 Dec 2025.

## Builds on (most cited first, 24 of 24)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer](https://mlsanity.com/paper/1907.01341/) (2019)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation](https://mlsanity.com/paper/1512.02134/) (2015)
- [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](https://mlsanity.com/paper/2301.00808/) (2023)
- [Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference](https://mlsanity.com/paper/1902.10556/) (2019)
- [Towards Real-Time Monocular Depth Estimation for Robotics: A Survey](https://mlsanity.com/paper/2111.08600/) (2021)
- [LRRU: Long-short Range Recurrent Updating Networks for Depth Completion](https://mlsanity.com/paper/2310.08956/) (2023)
- [Focus on defocus: bridging the synthetic to real domain gap for depth estimation](https://mlsanity.com/paper/2005.09623/) (2020)
- [Deep Depth From Focus](https://mlsanity.com/paper/1704.01085/) (2017)
- [Bridging Unsupervised and Supervised Depth from Focus via All-in-Focus Supervision](https://mlsanity.com/paper/2108.10843/) (2021)
- [A Survey on Deep Stereo Matching in the Twenties](https://mlsanity.com/paper/2407.07816/) (2024)
- [RM-Depth: Unsupervised Learning of Recurrent Monocular Depth in Dynamic Scenes](https://mlsanity.com/paper/2303.04456/) (2023)
- [Deep Depth from Focus with Differential Focus Volume](https://mlsanity.com/paper/2112.01712/) (2021)
- [Variational Depth from Focus Reconstruction](https://mlsanity.com/paper/1408.0173/) (2014)
- [Learning Depth from Focus in the Wild](https://mlsanity.com/paper/2207.09658/) (2022)
- [Deep Depth from Focal Stack with Defocus Model for Camera-Setting Invariance](https://mlsanity.com/paper/2202.13055/) (2022)
- [Active Depth Estimation: Stability Analysis and its Applications](https://mlsanity.com/paper/2003.07137/) (2020)
- [FocDepthFormer: Transformer with latent LSTM for Depth Estimation from Focal Stack](https://mlsanity.com/paper/2310.11178/) (2023)

## Built on by (most cited first, 1 of 1)

- [Towards Minimal Focal Stack in Shape from Focus](https://mlsanity.com/paper/2604.01603/) (2026)

## Abstract (arXiv)

Shape-from-Focus (SFF) is a passive depth estimation technique that infers scene depth by analyzing focus variations in a focal stack. Most recent deep learning-based SFF methods typically operate in two stages: first, they extract focus volumes (a per pixel representation of focus likelihood across the focal stack) using heavy feature encoders; then, they estimate depth via a simple one-step aggregation technique that often introduces artifacts and amplifies noise in the depth map. To address these issues, we propose a hybrid framework. Our method computes multi-scale focus volumes traditionally using handcrafted Directional Dilated Laplacian (DDL) kernels, which capture long-range and directional focus variations to form robust focus volumes. These focus volumes are then fed into a lightweight, multi-scale GRU-based depth extraction module that iteratively refines an initial depth estimate at a lower resolution for computational efficiency. Finally, a learned convex upsampling module within our recurrent network reconstructs high-resolution depth maps while preserving fine scene details and sharp boundaries. Extensive experiments on both synthetic and real-world datasets demonstrate that our approach outperforms state-of-the-art deep learning and traditional methods, achieving superior accuracy and generalization across diverse focal conditions.