Skip to content
RobotWorld
← Back

Google Gemini agentic video understanding

Loading video
Loading video

Google introduces agentic video understanding for Gemini, dynamically searching, scanning, and inspecting video content to improve long-video analysis while cutting tokens and cost.

GoogleGeminiAgentic video understanding智能体视频理解视频理解video understanding多模态AImultimodal AI多模态multimodal长视频long videoGorden_SunTwitter
Category: perception
Author: @Gorden_Sun
Date: 2026-09-02T00:00:00
Duration: 23.791s
Source: https://x.com/Gorden_Sun/status/2095011424020128059
Reference: https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/capabilities/video-understanding?hl=en

Related Videos

DeepMind D4RT turns a video into a 4D world
3:33

DeepMind D4RT turns a video into a 4D world

@Alacritic_Super · 94 views · 2026-08-31
D4RT4D世界4D world
VidMap: Exploiting temporal structure for video-based Structure-from-Motion
0:53

VidMap: Exploiting temporal structure for video-based Structure-from-Motion

@zador_pataki · 26 views · 2026-08-07
SLAMSfM三维重建
Ultralytics multi-object tracker comparison
0:16

Ultralytics multi-object tracker comparison

@Alacritic_Super · 48 views · 2026-09-02
目标跟踪多目标跟踪Ultralytics
Splat2Mesh: Free Tool Converting 3D Gaussian Splats to OBJ/GLB Meshes
0:07

Splat2Mesh: Free Tool Converting 3D Gaussian Splats to OBJ/GLB Meshes

@ymt3d · 27 views · 2026-09-01
3DGS高斯泼溅网格转换
InFlux++: benchmark and synthetic data for dynamic camera intrinsics
0:37

InFlux++: benchmark and synthetic data for dynamic camera intrinsics

@ErichLiang · 103 views · 2026-09-01
InFlux相机内参camera intrinsics
ABot-Recon turns driving video into 3D scenes
0:16

ABot-Recon turns driving video into 3D scenes

@HuggingApps · 75 views · 2026-08-31
ABot-Recon3D重建3D reconstruction