LongE2V: Long-Horizon Event-Based Video Reconstruction, Prediction, and Interpolation with Diffusion Priors

Forum topic · 小凯 · 2026-07-11

Summary

LongE2V (arXiv:2507.08182) is a computer vision paper by Cheng-De Fan, Chun-Wei Tuan Mu, and Chen-Wei Chang proposing a unified framework for recovering high-quality video from sparse event camera streams. It leverages pre-trained video diffusion priors to jointly handle event-based video reconstruction, prediction, and frame interpolation, achieving high data efficiency and superior perceptual quality through fine-tuning a foundational video model. To address temporal drift in extremely long sequences, the authors introduce Autoregressive Unrolling and Adaptive Context Switching. Reencoding Alignment with Cross Residual Correction ensures precise bidirectional consistency during frame interpolation, while Event Voxel Density Augmentation provides robustness across different sensor resolutions. Experiments on real-world benchmarks show LongE2V outperforms state-of-the-art methods on all three tasks, demonstrating strong temporal consistency and zero-shot generalization, overcoming the texture-blurring issues of regression methods and the instability of prior generative approaches.

Paper Overview

Research area: Computer Vision Authors: Cheng-De Fan, Chun-Wei Tuan Mu, Chen-Wei Chang arXiv: 2507.08182

Abstract

Recovering high-quality video from sparse event streams is a challenging task. Regression methods often blur textures, while existing generative models struggle with long-term stability.

The authors propose LongE2V, a novel approach that leverages pre-trained video diffusion priors to jointly handle event-based video reconstruction, prediction, and frame interpolation. By fine-tuning a foundational video model, the approach achieves high data efficiency and superior perceptual quality.

Key Contributions

Autoregressive Unrolling and Adaptive Context Switching: mitigates temporal drift in extremely long sequences.
Reencoding Alignment with Cross Residual Correction: ensures precise bidirectional consistency during frame interpolation.
Event Voxel Density Augmentation: ensures robustness to different event sensor resolutions.

Results

Extensive experiments on real-world benchmarks demonstrate that LongE2V outperforms state-of-the-art methods across all three tasks (reconstruction, prediction, and interpolation), exhibiting superior temporal consistency and zero-shot generalization.

--- *Source: forum post auto-collected 2026-07-11.*