# PX-NET: Simple and Efficient Pixel-Wise Training of Photometric Stereo Networks

arXiv: https://arxiv.org/abs/2008.04933 · HTML: https://mlsanity.com/paper/2008.04933/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Aug 2020.

## Built on by (most cited first, 21 of 21)

- [PS-NeRF: Neural Inverse Rendering for Multi-view Photometric Stereo](https://mlsanity.com/paper/2207.11406/) (2022)
- [S3-NeRF: Neural Reflectance Field from Shading and Shadow under a Single Viewpoint](https://mlsanity.com/paper/2210.08936/) (2022)
- [Scalable, Detailed and Mask-Free Universal Photometric Stereo](https://mlsanity.com/paper/2303.15724/) (2023)
- [Neural Reflectance for Shape Recovery with Shadow Handling](https://mlsanity.com/paper/2203.12909/) (2022)
- [Universal Photometric Stereo Network using Global Lighting Contexts](https://mlsanity.com/paper/2206.02452/) (2022)
- [Self-calibrating Photometric Stereo by Neural Inverse Rendering](https://mlsanity.com/paper/2207.07815/) (2022)
- [A CNN Based Approach for the Point-Light Photometric Stereo Problem](https://mlsanity.com/paper/2210.04655/) (2022)
- [Deep Learning Methods for Calibrated Photometric Stereo and Beyond](https://mlsanity.com/paper/2212.08414/) (2022)
- [Neural Architecture Search for Efficient Uncalibrated Deep Photometric Stereo](https://mlsanity.com/paper/2110.05621/) (2021)
- [DeepShadow: Neural Shape from Shadow](https://mlsanity.com/paper/2203.15065/) (2022)
- [Leveraging Spatial and Photometric Context for Calibrated Non-Lambertian Photometric Stereo](https://mlsanity.com/paper/2103.12106/) (2021)
- [Physics-Free Spectrally Multiplexed Photometric Stereo under Unknown Spectral Composition](https://mlsanity.com/paper/2410.20716/) (2024)
- [MS-PS: A Multi-Scale Network for Photometric Stereo With a New Comprehensive Training Dataset](https://mlsanity.com/paper/2211.14118/) (2022)
- [RMAFF-PSN: A Residual Multi-Scale Attention Feature Fusion Photometric Stereo Network](https://mlsanity.com/paper/2404.07766/) (2024)
- [Image Gradient-Aided Photometric Stereo Network](https://mlsanity.com/paper/2412.11650/) (2024)
- [Imperceptible Adversarial Examples for Fake Image Detection](https://mlsanity.com/paper/2106.01615/) (2021)
- [Open-World Object Detection via Discriminative Class Prototype Learning](https://mlsanity.com/paper/2302.11757/) (2023)
- [Event Fusion Photometric Stereo Network](https://mlsanity.com/paper/2303.00308/) (2023)
- [LIPIDS: Learning-based Illumination Planning In Discretized (Light) Space for Photometric Stereo](https://mlsanity.com/paper/2409.02716/) (2024)
- [PS-EIP: Robust Photometric Stereo Based on Event Interval Profile](https://mlsanity.com/paper/2503.18341/) (2025)
- [Event-based Photometric Stereo via Rotating Illumination and Per-Pixel Learning](https://mlsanity.com/paper/2603.10748/) (2026)

## Abstract (arXiv)

Retrieving accurate 3D reconstructions of objects from the way they reflect light is a very challenging task in computer vision. Despite more than four decades since the definition of the Photometric Stereo problem, most of the literature has had limited success when global illumination effects such as cast shadows, self-reflections and ambient light come into play, especially for specular surfaces. Recent approaches have leveraged the power of deep learning in conjunction with computer graphics in order to cope with the need of a vast number of training data in order to invert the image irradiance equation and retrieve the geometry of the object. However, rendering global illumination effects is a slow process which can limit the amount of training data that can be generated. In this work we propose a novel pixel-wise training procedure for normal prediction by replacing the training data (observation maps) of globally rendered images with independent per-pixel generated data. We show that global physical effects can be approximated on the observation map domain and this simplifies and speeds up the data creation procedure. Our network, PX-NET, achieves the state-of-the-art performance compared to other pixelwise methods on synthetic datasets, as well as the Diligent real dataset on both dense and sparse light settings.