E2Pano: Learning Event-to-Panorama Image Reconstruction

2026-08-01

AI Summary

This work addresses the limitations of existing event-based panoramic image reconstruction methods, which are either computationally expensive or struggle to model panoramic geometry accurately. We propose the first end-to-end learnable, geometry-guided reconstruction framework that preserves geometric consistency through spherical coordinate mapping. Our approach incorporates a lightweight frequency-domain supervision module and a spherical Transformer augmented with 3D positional encoding. Evaluated on both synthetic and real rotational scanning data, the method outperforms current baselines, achieving higher reconstruction quality and lower photometric cost. Notably, it successfully generalizes to real-world scenes despite being trained exclusively on synthetic data. To facilitate future research, we also introduce PanoScan, a new dataset comprising 4,370 synthetic and 30 real panoramic event sequences.

Abstract

Event cameras offer microsecond-level temporal resolution and high dynamic range, potentially facilitating motion-blur-free panoramic imaging from fast rotational scanning. Nonetheless, existing optimization-based methods remain computationally demanding, while prior learning-based reconstruction methods are largely designed for perspective imagery and lack geometry-aware support for panoramic outputs. We present E2Pano, a geometry-guided event-to-panorama pipeline with an end-to-end learnable photometric reconstruction stage. Our framework preserves real spherical coordinates from geometric mapping throughout the pipeline, employs a lightweight enhancement module with frequency-domain supervision to bridge the event-image domain gap, and leverages a spherical Transformer with 3D positional embeddings for photometric reconstruction. Experiments on synthetic data and captured rotational scans show improved reconstruction quality and lower photometric reconstruction cost than optimization-based baselines, together with encouraging transfer to real captures under our acquisition protocol despite training purely on synthetic data. Additionally, we construct PanoScan, a dataset with 4,370 synthetic and 30 real-world panoramic scenes paired with event streams. Our dataset and code will be released.

Problem

• event camera
• panoramic imaging
• image reconstruction
• geometry-aware
• rotational scanning

Innovation

• event camera
• panoramic imaging
• geometry-aware reconstruction
• spherical Transformer
• frequency-domain supervision