# PanoNormal: Monocular Indoor 360deg Surface Normal Estimation

arXiv: https://arxiv.org/abs/2405.18745 · HTML: https://mlsanity.com/paper/2405.18745/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 41 references that are in the map.
- First posted on arXiv 29 May 2024.

## Builds on (most cited first, 29 of 29)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Matterport3D: Learning from RGB-D Data in Indoor Environments](https://mlsanity.com/paper/1709.06158/) (2017)
- [Uformer: A General U-Shaped Transformer for Image Restoration](https://mlsanity.com/paper/2106.03106/) (2021)
- [Semantic Scene Completion from a Single Depth Image](https://mlsanity.com/paper/1611.08974/) (2016)
- [Joint 2D-3D-Semantic Data for Indoor Scene Understanding](https://mlsanity.com/paper/1702.01105/) (2017)
- [Structured3D: A Large Photo-realistic Dataset for Structured 3D Modeling](https://mlsanity.com/paper/1908.00222/) (2019)
- [Incorporating Convolution Designs into Visual Transformers](https://mlsanity.com/paper/2103.11816/) (2021)
- [BiFormer: Vision Transformer with Bi-Level Routing Attention](https://mlsanity.com/paper/2303.08810/) (2023)
- [MonoViT: Self-Supervised Monocular Depth Estimation with a Vision Transformer](https://mlsanity.com/paper/2208.03543/) (2022)
- [Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation](https://mlsanity.com/paper/2109.09881/) (2021)
- [OmniDepth: Dense Depth Estimation for Indoors Spherical Panoramas](https://mlsanity.com/paper/1807.09620/) (2018)
- [PanoFormer: Panorama Transformer for Indoor 360 Depth Estimation](https://mlsanity.com/paper/2203.09283/) (2022)
- [UniFuse: Unidirectional Fusion for 360circ Panorama Depth Estimation](https://mlsanity.com/paper/2102.03550/) (2021)
- [Bending Reality: Distortion-aware Transformers for Adapting to Panoramic Semantic Segmentation](https://mlsanity.com/paper/2203.01452/) (2022)
- [OmniFusion: 360 Monocular Depth Estimation via Geometry-Aware Fusion](https://mlsanity.com/paper/2203.00838/) (2022)
- [SeMask: Semantically Masked Transformers for Semantic Segmentation](https://mlsanity.com/paper/2112.12782/) (2021)
- [EGformer: Equirectangular Geometry-biased Transformer for 360 Depth Estimation](https://mlsanity.com/paper/2304.07803/) (2023)
- [Spherical Regression: Learning Viewpoints, Surface Normals and 3D Rotations on n-Spheres](https://mlsanity.com/paper/1904.05404/) (2019)
- [Pano3D: A Holistic Benchmark and a Solid Baseline for 360o Depth Estimation](https://mlsanity.com/paper/2109.02749/) (2021)
- [HRDFuse: Monocular 360degDepth Estimation by Collaboratively Learning Holistic-with-Regional Depth Distributions](https://mlsanity.com/paper/2303.11616/) (2023)
- [BiFuse++: Self-supervised and Efficient Bi-projection Fusion for 360 Depth Estimation](https://mlsanity.com/paper/2209.02952/) (2022)
- [Surface Normal Estimation of Tilted Images via Spatial Rectifier](https://mlsanity.com/paper/2007.09264/) (2020)
- [Elite360D: Towards Efficient 360 Depth Estimation via Semantic- and Distance-Aware Bi-Projection Fusion](https://mlsanity.com/paper/2403.16376/) (2024)
- [GLPanoDepth: Global-to-Local Panoramic Depth Estimation](https://mlsanity.com/paper/2202.02796/) (2022)
- [Panoramic Vision Transformer for Saliency Detection in 360deg Videos](https://mlsanity.com/paper/2209.08956/) (2022)
- [Adaptive Surface Normal Constraint for Geometric Estimation from Monocular Images](https://mlsanity.com/paper/2402.05869/) (2024)
- [SGFormer: Spherical Geometry Transformer for 360 Depth Estimation](https://mlsanity.com/paper/2404.14979/) (2024)

## Built on by (most cited first, 6 of 6)

- [MTPano: Multi-Task Panoramic Scene Understanding via Label-Free Integration of Dense Prediction Priors](https://mlsanity.com/paper/2602.05330/) (2026)
- [OB3D: A New Dataset for Benchmarking Omnidirectional 3D Reconstruction Using Blender](https://mlsanity.com/paper/2505.20126/) (2025)
- [Unified Panoramic Geometry Estimation via Multi-View Foundation Models](https://mlsanity.com/paper/2605.26368/) (2026)
- [Panoramic Scene Understanding: A Survey from Distortion-Aware Engineering to Sphere-Native Modeling](https://mlsanity.com/paper/2606.27745/) (2026)
- [Cross360: 360deg Monocular Depth Estimation via Cross Projections Across Scales](https://mlsanity.com/paper/2601.17271/) (2026)
- [H-OmniStereo: Zero-Shot Omnidirectional Stereo Matching with Heading-Aligned Normal Priors](https://mlsanity.com/paper/2605.14963/) (2026)

## Abstract (arXiv)

The presence of spherical distortion in equirectangular projection (ERP) images presents a persistent challenge in dense regression tasks such as surface normal estimation. Although it may appear straightforward to repurpose architectures developed for 360deg depth estimation, our empirical findings indicate that such models yield suboptimal performance when applied to surface normal prediction. This is largely attributed to their architectural bias toward capturing global scene layout, which comes at the expense of the fine-grained local geometric cues that are critical for accurate surface orientation estimation. While convolutional neural networks (CNNs) have been employed to mitigate spherical distortion, their fixed receptive fields limit their ability to capture holistic scene structure. Conversely, vision transformers (ViTs) are capable of modeling long-range dependencies via global self-attention, but often fail to preserve high-frequency local detail. To address these limitations, we propose textitPanoNormal, a monocular surface normal estimation architecture for 360deg images that integrates the complementary strengths of CNNs and ViTs. In particular, we design a multi-level global self-attention mechanism that explicitly accounts for the spherical feature distribution, enabling our model to recover both global contextual structure and local geometric details. Experimental results demonstrate that our method not only achieves state-of-the-art performance on several benchmark 360deg datasets, but also significantly outperforms adapted depth estimation models on the task of surface normal prediction. The code and model are available at https://github.com/huangkun101230/PanoNormal.