3260 papers • 126 benchmarks • 313 datasets
Event-Based Video Reconstruction aims to generate a sequence of intensity frames from an asynchronous stream of events (per-pixel brightness change signals outputted by an event camera).
(Image credit: Papersgraph)
These leaderboards are used to track progress in event-based-video-reconstruction-11
Use these libraries to find event-based-video-reconstruction-11 models and implementations
No subtasks available.
This work presents a new High Quality Frames (HQF) dataset, containing events and ground truth frames from a DAVIS240C that are well-exposed and minimally motion-blurred, and presents strategies for improving training data for event based CNNs that result in 20-40% boost in performance of existing state-of-the-art (SOTA) video reconstruction networks retrained with this method.
The proposed algorithm includes a frame augmentation pre-processing step that deblurs and temporally interpolates frame data using events and outperforms state-of-the-art methods in both absolute intensity error and image similarity indexes.
Event cameras, which output events by detecting spatio- temporal brightness changes, bring a novel paradigm to image sensors with high dynamic range and low latency. Previous works have achieved impressive performances on event-based video reconstruction by introducing convolutional neural networks (CNNs). However, intrinsic locality of convolutional operations is not capable of modeling long-range dependency, which is crucial to many vision tasks. In this paper, we present a hybrid CNN- Transformer network for event-based video reconstruction (ET-Net), which merits the fine local information from CNN and global contexts from Transformer In addition, we further propose a Token Pyramid Aggregation strategy to implement multi-scale token integration for relating internal and intersected semantic concepts in the token-space. Experimental results demonstrate that our proposed method achieves superior performance over state-of-the-art methods on multiple real-world event datasets. The code is available at https://github.com/WarranWeng/ET-Net.
This paper proposes a novel Event-based Video reconstruction framework based on a fully Spiking Neural Network (EVSNN), which utilizes Leaky-Integrate-and-Fire (LIF) neuron and Membrane Potential (MP) neuron, and finds that the spiking neurons have the potential to store useful temporal information (memory) to complete such time-dependent tasks.
This paper proposes a unified evaluation methodology and introduces an open-source framework called EVREAL to comprehensively benchmark and analyze various event-based video reconstruction methods from the literature and provides valuable insights into the performance of these methods under varying settings, challenging scenarios, and downstream tasks.
This study proposes HyperE2VID, a dynamic neural network architecture for event-based video reconstruction that uses hypernetworks to generate per-pixel adaptive filters guided by a context fusion module that combines information from event voxel grids and previously reconstructed intensity images.
Adding a benchmark result helps the community track progress.