# Kernel Transformer Networks for Compact Spherical Convolution

arXiv: https://arxiv.org/abs/1812.03115 · HTML: https://mlsanity.com/paper/1812.03115/

## Key facts

- Cited by 56 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Dec 2018.

## Built on by (most cited first, 30 of 56)

- [PanoFormer: Panorama Transformer for Indoor 360 Depth Estimation](https://mlsanity.com/paper/2203.09283/) (2022)
- [OmniFusion: 360 Monocular Depth Estimation via Geometry-Aware Fusion](https://mlsanity.com/paper/2203.00838/) (2022)
- [360MonoDepth: High-Resolution 360deg Monocular Depth Estimation](https://mlsanity.com/paper/2111.15669/) (2021)
- [Rethinking Spatial Invariance of Convolutional Networks for Object Counting](https://mlsanity.com/paper/2206.05253/) (2022)
- [Both Style and Distortion Matter: Dual-Path Unsupervised Domain Adaptation for Panoramic Semantic Segmentation](https://mlsanity.com/paper/2303.14360/) (2023)
- [LED2-Net: Monocular 360 Layout Estimation via Differentiable Depth Rendering](https://mlsanity.com/paper/2104.00568/) (2021)
- [Deep Learning for Omnidirectional Vision: A Survey and New Perspectives](https://mlsanity.com/paper/2205.10468/) (2022)
- [Depth Anywhere: Enhancing 360 Monocular Depth Estimation via Perspective Distillation and Unlabeled Data Augmentation](https://mlsanity.com/paper/2406.12849/) (2024)
- [BiFuse++: Self-supervised and Efficient Bi-projection Fusion for 360 Depth Estimation](https://mlsanity.com/paper/2209.02952/) (2022)
- [Surround-view Fisheye Camera Perception for Automated Driving: Overview, Survey and Challenges](https://mlsanity.com/paper/2205.13281/) (2022)
- [SphereSR: 360deg Image Super-Resolution with Arbitrary Projection via Continuous Spherical Image Representation](https://mlsanity.com/paper/2112.06536/) (2021)
- [Elite360D: Towards Efficient 360 Depth Estimation via Semantic- and Distance-Aware Bi-Projection Fusion](https://mlsanity.com/paper/2403.16376/) (2024)
- [A Survey of Representation Learning, Optimization Strategies, and Applications for Omnidirectional Vision](https://mlsanity.com/paper/2502.10444/) (2025)
- [Scaling Spherical CNNs](https://mlsanity.com/paper/2306.05420/) (2023)
- [PanoSwin: a Pano-style Swin Transformer for Panorama Understanding](https://mlsanity.com/paper/2308.14726/) (2023)
- [GLPanoDepth: Global-to-Local Panoramic Depth Estimation](https://mlsanity.com/paper/2202.02796/) (2022)
- [360 Depth Estimation in the Wild -- The Depth360 Dataset and the SegFuse Network](https://mlsanity.com/paper/2202.08010/) (2022)
- [Panoramic Vision Transformer for Saliency Detection in 360deg Videos](https://mlsanity.com/paper/2209.08956/) (2022)
- [PANORAMA: The Rise of Omnidirectional Vision in the Embodied AI Era](https://mlsanity.com/paper/2509.12989/) (2025)
- [Spherical Transformer](https://mlsanity.com/paper/2202.04942/) (2022)
- [FisheyeHDK: Hyperbolic Deformable Kernel Learning for Ultra-Wide Field-of-View Image Recognition](https://mlsanity.com/paper/2203.07255/) (2022)
- [DarSwin: Distortion Aware Radial Swin Transformer](https://mlsanity.com/paper/2304.09691/) (2023)
- [PanoDepth: A Two-Stage Approach for Monocular Omnidirectional Depth Estimation](https://mlsanity.com/paper/2202.01323/) (2022)
- [SGFormer: Spherical Geometry Transformer for 360 Depth Estimation](https://mlsanity.com/paper/2404.14979/) (2024)
- [Deep 360^circ Optical Flow Estimation Based on Multi-Projection Fusion](https://mlsanity.com/paper/2208.00776/) (2022)
- [FoVA-Depth: Field-of-View Agnostic Depth Estimation for Cross-Dataset Generalization](https://mlsanity.com/paper/2401.13786/) (2024)
- [Single-Shot Cuboids: Geodesics-based End-to-end Manhattan Aligned Layout Estimation from Spherical Panoramas](https://mlsanity.com/paper/2102.03939/) (2021)
- [360deg Optical Flow using Tangent Images](https://mlsanity.com/paper/2112.14331/) (2021)
- [Distortion-Tolerant Monocular Depth Estimation On Omnidirectional Images Using Dual-cubemap](https://mlsanity.com/paper/2203.09733/) (2022)
- [3D reconstruction from spherical images: A review of techniques, applications, and prospects](https://mlsanity.com/paper/2302.04495/) (2023)

## Abstract (arXiv)

Ideally, 360deg imagery could inherit the deep convolutional neural networks (CNNs) already trained with great success on perspective projection images. However, existing methods to transfer CNNs from perspective to spherical images introduce significant computational costs and/or degradations in accuracy. In this work, we present the Kernel Transformer Network (KTN). KTNs efficiently transfer convolution kernels from perspective images to the equirectangular projection of 360deg images. Given a source CNN for perspective images as input, the KTN produces a function parameterized by a polar angle and kernel as output. Given a novel 360deg image, that function in turn can compute convolutions for arbitrary layers and kernels as would the source CNN on the corresponding tangent plane projections. Distinct from all existing methods, KTNs allow model transfer: the same model can be applied to different source CNNs with the same base architecture. This enables application to multiple recognition tasks without re-training the KTN. Validating our approach with multiple source CNNs and datasets, we show that KTNs improve the state of the art for spherical convolution. KTNs successfully preserve the source CNN's accuracy, while offering transferability, scalability to typical image resolutions, and, in many cases, a substantially lower memory footprint.