# Occlusion-aware Unsupervised Learning of Depth from 4-D Light Fields

arXiv: https://arxiv.org/abs/2106.03043 · HTML: https://mlsanity.com/paper/2106.03043/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 49 references that are in the map.
- First posted on arXiv 6 Jun 2021.

## Builds on (most cited first, 12 of 12)

- [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)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [FlowNet: Learning Optical Flow with Convolutional Networks](https://mlsanity.com/paper/1504.06852/) (2015)
- [FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks](https://mlsanity.com/paper/1612.01925/) (2016)
- [Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement](https://mlsanity.com/paper/2001.06826/) (2020)
- [Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild](https://mlsanity.com/paper/1911.11130/) (2019)
- [Learning-Based View Synthesis for Light Field Cameras](https://mlsanity.com/paper/1609.02974/) (2016)
- [What Matters in Unsupervised Optical Flow](https://mlsanity.com/paper/2006.04902/) (2020)
- [EPINET: A Fully-Convolutional Neural Network Using Epipolar Geometry for Depth from Light Field Images](https://mlsanity.com/paper/1804.02379/) (2018)
- [Accurate Light Field Depth Estimation with Superpixel Regularization over Partially Occluded Regions](https://mlsanity.com/paper/1708.01964/) (2017)
- [Occlusion-Model Guided Anti-Occlusion Depth Estimation in Light Field](https://mlsanity.com/paper/1608.04187/) (2016)

## Built on by (most cited first, 16 of 16)

- [Occlusion-Aware Cost Constructor for Light Field Depth Estimation](https://mlsanity.com/paper/2203.01576/) (2022)
- [Learning Non-Local Spatial-Angular Correlation for Light Field Image Super-Resolution](https://mlsanity.com/paper/2302.08058/) (2023)
- [LRT: An Efficient Low-Light Restoration Transformer for Dark Light Field Images](https://mlsanity.com/paper/2209.02197/) (2022)
- [OPAL: Occlusion Pattern Aware Loss for Unsupervised Light Field Disparity Estimation](https://mlsanity.com/paper/2203.02231/) (2022)
- [Probabilistic-based Feature Embedding of 4-D Light Fields for Compressive Imaging and Denoising](https://mlsanity.com/paper/2306.08836/) (2023)
- [Light Field Depth Estimation via Stitched Epipolar Plane Images](https://mlsanity.com/paper/2203.15201/) (2022)
- [Unsupervised Light Field Depth Estimation via Multi-view Feature Matching with Occlusion Prediction](https://mlsanity.com/paper/2301.08433/) (2023)
- [Blur aware metric depth estimation with multi-focus plenoptic cameras](https://mlsanity.com/paper/2308.04252/) (2023)
- [Enhancing Underwater Imaging with 4-D Light Fields: Dataset and Method](https://mlsanity.com/paper/2408.17339/) (2024)
- [Learning based Deep Disentangling Light Field Reconstruction and Disparity Estimation Application](https://mlsanity.com/paper/2311.08129/) (2023)
- [Self-supervised Learning-based Reconstruction of High-resolution 4D Light Fields](https://mlsanity.com/paper/2402.19020/) (2024)
- [Iterative Occlusion-Aware Light Field Depth Estimation using 4D Geometrical Cues](https://mlsanity.com/paper/2403.02043/) (2024)
- [DSER: Spectral Epipolar Representation for Efficient Light Field Depth Estimation](https://mlsanity.com/paper/2508.08900/) (2025)
- [Exploring Non-Local Spatial-Angular Correlations with a Hybrid Mamba-Transformer Framework for Light Field Super-Resolution](https://mlsanity.com/paper/2509.04824/) (2025)
- [Enhancing Underwater Light Field Images via Global Geometry-aware Diffusion Process](https://mlsanity.com/paper/2601.21179/) (2026)
- [UniqueSplat: View-conditioned 3D Gaussian Splatting for Generalizable 3D Reconstruction](https://mlsanity.com/paper/2608.02145/) (2026)

## Abstract (arXiv)

Depth estimation is a fundamental issue in 4-D light field processing and analysis. Although recent supervised learning-based light field depth estimation methods have significantly improved the accuracy and efficiency of traditional optimization-based ones, these methods rely on the training over light field data with ground-truth depth maps which are challenging to obtain or even unavailable for real-world light field data. Besides, due to the inevitable gap (or domain difference) between real-world and synthetic data, they may suffer from serious performance degradation when generalizing the models trained with synthetic data to real-world data. By contrast, we propose an unsupervised learning-based method, which does not require ground-truth depth as supervision during training. Specifically, based on the basic knowledge of the unique geometry structure of light field data, we present an occlusion-aware strategy to improve the accuracy on occlusion areas, in which we explore the angular coherence among subsets of the light field views to estimate initial depth maps, and utilize a constrained unsupervised loss to learn their corresponding reliability for final depth prediction. Additionally, we adopt a multi-scale network with a weighted smoothness loss to handle the textureless areas. Experimental results on synthetic data show that our method can significantly shrink the performance gap between the previous unsupervised method and supervised ones, and produce depth maps with comparable accuracy to traditional methods with obviously reduced computational cost. Moreover, experiments on real-world datasets show that our method can avoid the domain shift problem presented in supervised methods, demonstrating the great potential of our method.