# Omnidirectional CNN for Visual Place Recognition and Navigation

arXiv: https://arxiv.org/abs/1803.04228 · HTML: https://mlsanity.com/paper/1803.04228/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Mar 2018.

## Built on by (most cited first, 22 of 22)

- [GSV-Cities: Toward Appropriate Supervised Visual Place Recognition](https://mlsanity.com/paper/2210.10239/) (2022)
- [UniFuse: Unidirectional Fusion for 360circ Panorama Depth Estimation](https://mlsanity.com/paper/2102.03550/) (2021)
- [ACDNet: Adaptively Combined Dilated Convolution for Monocular Panorama Depth Estimation](https://mlsanity.com/paper/2112.14440/) (2021)
- [BiFuse++: Self-supervised and Efficient Bi-projection Fusion for 360 Depth Estimation](https://mlsanity.com/paper/2209.02952/) (2022)
- [Robust Knowledge Graph Completion with Stacked Convolutions and a Student Re-Ranking Network](https://mlsanity.com/paper/2106.06555/) (2021)
- [3D Scene Geometry Estimation from 360^circ Imagery: A Survey](https://mlsanity.com/paper/2401.09252/) (2024)
- [Indoor Panorama Planar 3D Reconstruction via Divide and Conquer](https://mlsanity.com/paper/2106.14166/) (2021)
- [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)
- [HyCubE: Efficient Knowledge Hypergraph 3D Circular Convolutional Embedding](https://mlsanity.com/paper/2402.08961/) (2024)
- [PanoVPR: Towards Unified Perspective-to-Equirectangular Visual Place Recognition via Sliding Windows across the Panoramic View](https://mlsanity.com/paper/2303.14095/) (2023)
- [Context-aware Padding for Semantic Segmentation](https://mlsanity.com/paper/2109.07854/) (2021)
- [PRFusion: Toward Effective and Robust Multi-Modal Place Recognition with Image and Point Cloud Fusion](https://mlsanity.com/paper/2410.04939/) (2024)
- [Depth Anything in 360^circ: Towards Scale Invariance in the Wild](https://mlsanity.com/paper/2512.22819/) (2025)
- [Panoramic Image-to-Image Translation](https://mlsanity.com/paper/2304.04960/) (2023)
- [Docmarking: Real-Time Screen-Cam Robust Document Image Watermarking](https://mlsanity.com/paper/2304.12682/) (2023)
- [Boosting Omnidirectional Stereo Matching with a Pre-trained Depth Foundation Model](https://mlsanity.com/paper/2503.23502/) (2025)
- [Deep Spherical Superpixels](https://mlsanity.com/paper/2407.17354/) (2024)
- [HypeVPR: Exploring Hyperbolic Space for Perspective to Equirectangular Visual Place Recognition](https://mlsanity.com/paper/2506.04764/) (2025)
- [DAOVI: Distortion-Aware Omnidirectional Video Inpainting](https://mlsanity.com/paper/2509.00396/) (2025)
- [EdgeLDR: Quaternion Low-Displacement Rank Neural Networks for Edge-Efficient Deep Learning](https://mlsanity.com/paper/2601.05379/) (2026)
- [SphMind: Towards Robust, Training-Free VLM-based Spatial Reasoning with a 360 Camera](https://mlsanity.com/paper/2609.33462/) (2026)

## Abstract (arXiv)

 Visual place recognition is challenging, especially when only a few place exemplars are given. To mitigate the challenge, we consider place recognition method using omnidirectional cameras and propose a novel Omnidirectional Convolutional Neural Network (O-CNN) to handle severe camera pose variation. Given a visual input, the task of the O-CNN is not to retrieve the matched place exemplar, but to retrieve the closest place exemplar and estimate the relative distance between the input and the closest place. With the ability to estimate relative distance, a heuristic policy is proposed to navigate a robot to the retrieved closest place. Note that the network is designed to take advantage of the omnidirectional view by incorporating circular padding and rotation invariance. To train a powerful O-CNN, we build a virtual world for training on a large scale. We also propose a continuous lifted structured feature embedding loss to learn the concept of distance efficiently. Finally, our experimental results confirm that our method achieves state-of-the-art accuracy and speed with both the virtual world and real-world datasets.