AI Paper Atlas

Research map / Video / Video segmentation and tracking

EventDrive: Event Cameras for Vision-Language Driving Intelligence

Dongyue Lu, Rong Li, Ao Liang, Lingdong Kong, Wei Yin, Lai Xing Ng, Benoit R. Cottereau, Camille Simon Chane, Wei Tsang Ooi

CVPR 2026 · Video › Video segmentation and tracking · approach: Event-based video reconstruction · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add event stream to frame inputQwen2.5-VL-7B fine-tuned · EventDrive PlanningL2 error: 4.54 -> 3.66Improved
add event stream to frame inputQwen2.5-VL-7B fine-tuned · EventDrive groundingmIoU: 67.19 -> 72.56Improved
add event stream to frame inputQwen2.5-VL-7B fine-tuned · EventDrive Prediction speedaccuracy: 36.84 -> 42.44Improved
add event stream to frame inputQwen2.5-VL-7B fine-tuned · EventDrive UnderstandingQA accuracy: 58.44 -> 65.46Improved
fine-tune on EventDrive instead of zero-shotQwen2.5-VL-7B-Instruct · EventDrive PlanningL2 error: 8.78 -> 4.54Improved
use event-only input instead of EventGPT-7B zero-shotEventDrive-VLM · EventDrive PlanningL2 error: 11.42 -> 6.89Improved
increase horizons from 1 to 3 (ablation)EventDrive-VLM · EventDrive perceptionQA accuracy: 82.40 -> 83.66Improved
increase horizons from 3 to 5 (ablation)EventDrive-VLM · EventDrive perceptionQA accuracy: 83.66 -> 83.95No clear effect
use MoE top-1 expert instead of Add aggregation (ablation)EventDrive-VLM · EventDrive perceptionQA accuracy: 76.76 -> 83.66Improved
use MoE top-1 expert instead of weighted sum (ablation)EventDrive-VLM · EventDrive predictionaccuracy: 61.08 -> 64.96Improved
use Event Q-Former instead of concatenation (ablation)EventDrive-VLM · EventDrive perceptionQA accuracy: 79.35 -> 83.66Improved
use Event Q-Former instead of cross-attention (ablation)EventDrive-VLM · EventDrive planningL2 error: 3.69 -> 3.66Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy