# Event ActivityNet: A Large-Scale Simulated-Event Benchmark for Untrimmed Action Understanding

arXiv: https://arxiv.org/abs/2608.01948 · HTML: https://mlsanity.com/paper/2608.01948/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 30 references that are in the map.
- First posted on arXiv 3 Aug 2026.

## Builds on (most cited first, 17 of 17)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Scaling Egocentric Vision: The EPIC-KITCHENS Dataset](https://mlsanity.com/paper/1804.02748/) (2018)
- [Dense-Captioning Events in Videos](https://mlsanity.com/paper/1705.00754/) (2017)
- [Unsupervised Event-based Learning of Optical Flow, Depth, and Egomotion](https://mlsanity.com/paper/1812.08156/) (2018)
- [The THUMOS Challenge on Action Recognition for Videos "in the Wild"](https://mlsanity.com/paper/1604.06182/) (2016)
- [Events-to-Video: Bringing Modern Computer Vision to Event Cameras](https://mlsanity.com/paper/1904.08298/) (2019)
- [Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining](https://mlsanity.com/paper/2302.02318/) (2023)
- [Combining Events and Frames using Recurrent Asynchronous Multimodal Networks for Monocular Depth Prediction](https://mlsanity.com/paper/2102.09320/) (2021)
- [GET: Group Event Transformer for Event-Based Vision](https://mlsanity.com/paper/2310.02642/) (2023)
- [HyperE2VID: Improving Event-Based Video Reconstruction via Hypernetworks](https://mlsanity.com/paper/2305.06382/) (2023)
- [MECCANO: A Multimodal Egocentric Dataset for Humans Behavior Understanding in the Industrial-like Domain](https://mlsanity.com/paper/2209.08691/) (2022)
- [ExACT: Language-guided Conceptual Reasoning and Uncertainty Estimation for Event-based Action Recognition and More](https://mlsanity.com/paper/2403.12534/) (2024)
- [V2CE: Video to Continuous Events Simulator](https://mlsanity.com/paper/2309.08891/) (2023)
- [Hypergraph-based Multi-View Action Recognition using Event Cameras](https://mlsanity.com/paper/2403.19316/) (2024)
- [V2V: Scaling Event-Based Vision through Efficient Video-to-Voxel Simulation](https://mlsanity.com/paper/2505.16797/) (2025)
- [PreFM: Online Audio-Visual Event Parsing via Predictive Future Modeling](https://mlsanity.com/paper/2505.23155/) (2025)
- [EventFlash: Towards Efficient MLLMs for Event-Based Vision](https://mlsanity.com/paper/2602.03230/) (2026)

## Abstract (arXiv)

Long-horizon event-based action understanding remains underexplored because existing datasets largely comprise short, trimmed clips, while collecting native event streams with dense temporal annotations is costly. We introduce Event ActivityNet, a large-scale simulated-event benchmark derived from human-annotated, untrimmed ActivityNet videos. It comprises 3,263 videos, 200 action classes, and 106.94 hours, with matched 5-bin and 9-bin event-voxel representations, temporal action annotations, and timestamped captions. The benchmark supports annotated-segment action recognition, auxiliary event-language alignment, and causal online temporal action localization. We generate event voxels directly from non-interpolated source videos in decoded frame order, retain per-video rational nominal or average frame-rate metadata for approximate time mapping, and use action-center reconstruction LPIPS as a soft diagnostic of retained reconstructable content. We establish baselines for adaptive event framing, prompt-caption alignment, and event-only, RGB-only, and RGB-event localization. Under a progressive nested-scale training protocol, recognition Top-1 accuracy increases from 52.25 to 66.42, while online temporal localization average mAP improves from 21.7 to 29.0. Moreover, staged Event ActivityNet pretraining followed by native-event fine-tuning consistently outperforms target-only and joint-from-scratch training across multiple supervision budgets. Event ActivityNet provides a scalable benchmark for long-horizon event modeling, although native-camera evaluation remains essential for deployment-oriented conclusions.