Sector Patch Embedding: An Embedding Module Conforming to The Distortion Pattern of Fisheye Image

TL;DR

This paper introduces Sector Patch Embedding (SPE), a novel embedding method tailored to fisheye image distortion patterns, improving Transformer model performance on fisheye image classification tasks.

Contribution

The paper proposes SPE, a new patch embedding technique that aligns with fisheye distortion patterns, enhancing feature extraction and model accuracy.

Findings

SPE improves ViT top-1 accuracy by 0.75%.

SPE improves PVT top-1 accuracy by 2.8%.

The method effectively perceives distortion in fisheye images.

Abstract

Fisheye cameras suffer from image distortion while having a large field of view(LFOV). And this fact leads to poor performance on some fisheye vision tasks. One of the solutions is to optimize the current vision algorithm for fisheye images. However, most of the CNN-based methods and the Transformer-based methods lack the capability of leveraging distortion information efficiently. In this work, we propose a novel patch embedding method called Sector Patch Embedding(SPE), conforming to the distortion pattern of the fisheye image. Furthermore, we put forward a synthetic fisheye dataset based on the ImageNet-1K and explore the performance of several Transformer models on the dataset. The classification top-1 accuracy of ViT and PVT is improved by 0.75% and 2.8% with SPE respectively. The experiments show that the proposed sector patch embedding method can better perceive distortion and extract features on the fisheye images. Our method can be easily adopted to other Transformer-based models. Source code is at https://github.com/IN2-ViAUn/Sector-Patch-Embedding.