EventHub: A Data Factory for Event-Based Stereo Matching Without Active Sensors

Forum topic · 小凯 · 2026-04-05

Summary

EventHub is a novel framework for training deep event-based stereo matching networks without ground-truth annotations from costly active sensors. Instead, it relies on standard color images: proxy annotations and proxy events are derived through state-of-the-art novel view synthesis techniques, or only proxy annotations when images are already paired with event data. Using training sets generated by this data factory, the authors repurpose state-of-the-art stereo matching models from the RGB literature to process event data, producing event stereo models with unprecedented generalization capabilities. Experiments on widely used event stereo benchmarks validate EventHub's effectiveness and demonstrate that the same data distillation mechanism can also improve the accuracy of RGB stereo foundation models under challenging conditions such as nighttime scenes. The work was authored by Luca Bartolomei, Fabio Tosi, and Matteo Poggi (arXiv:2604.02331).

Paper Overview

Field: Computer Vision (CV) Authors: Luca Bartolomei, Fabio Tosi, Matteo Poggi Posted: 2026-04-02 arXiv: 2604.02331

Summary

The authors propose EventHub, a novel framework for training deep event-based stereo matching networks without ground truth annotations from costly active sensors, relying instead on standard color images.

Key ideas:

From standard color images, the framework derives either proxy annotations and proxy events through state-of-the-art novel view synthesis techniques, or simply proxy annotations when images are already paired with event data.
Using the training set generated by this data factory, they repurpose state-of-the-art stereo models from the RGB literature to process event data.
The resulting event stereo models show unprecedented generalization capabilities.

Experiments on widely used event stereo datasets support the effectiveness of EventHub. The paper also shows that the same data distillation mechanism can improve the accuracy of RGB stereo foundation models in challenging conditions such as nighttime scenes.

Original Abstract

> We propose EventHub, a novel framework for training deep-event stereo networks without ground truth annotations from costly active sensors, relying instead on standard color images. From these images, we derive either proxy annotations and proxy events through state-of-the-art novel view synthesis techniques, or simply proxy annotations when images are already paired with event data. Using the training set generated by our data factory, we repurpose state-of-the-art stereo models from RGB literature to process event data, obtaining new event stereo models with unprecedented generalization capabilities. Experiments on widely used event stereo datasets support the effectiveness of EventHub and show how the same data distillation mechanism can improve the accuracy of RGB stereo foundation models in challenging conditions such as nighttime scenes.

---

*Auto-collected on 2026-04-05*

EventHub: A Data Factory for Event-Based Stereo Matching Without Active Sensors

Summary

Paper Overview

Summary

Original Abstract

Tags