Skip to content
RobotWorld
← Back

Inside SuperMap: Persistent 3D Object Identities

Loading video
Loading video

SuperOdometry builds the geometric map, GroundingDINO and SAM2 provide open-vocabulary detections, and a spatio-temporal association layer turns noisy detections into persistent 3D object identities.

SLAMSuperMapSuperOdometryGroundingDINOSAM2开放词汇open-vocabulary3D物体感知perceptionTwitter
Category: perception
Author: @robotsdigest
Date: 2026-08-16T00:00:00
Duration: 36.286s
Source: https://x.com/robotsdigest/status/2089052774453620833

Related Videos

SuperMap Turns SLAM into Persistent 4D Spatial Memory
3:52

SuperMap Turns SLAM into Persistent 4D Spatial Memory

@robotsdigest · 6 views · 2026-08-16
SLAMSuperMap空间记忆
SuperMap: Spatio-Temporal SLAM for Visual-Language Navigation
1:00

SuperMap: Spatio-Temporal SLAM for Visual-Language Navigation

@ShiboZhaoSLAM · 4 views · 2026-07-14
SLAMSuperMap视觉语言导航
PRISM-VO: Scale-Aware Visual Odometry via Photometric Plenoptic Bundle Adjustment
0:37

PRISM-VO: Scale-Aware Visual Odometry via Photometric Plenoptic Bundle Adjustment

@Almorgand · 3 views · 2026-07-29
视觉里程计visual odometry光场相机
REMIND Re-ID Tracker for Indoor Navigation
0:22

REMIND Re-ID Tracker for Indoor Navigation

@rsasaki0109 · 3 views · 2026-07-30
重识别re-identification视觉跟踪
Monocular Camera Autonomous Driving Perception
0:13

Monocular Camera Autonomous Driving Perception

@tokufxug · 4 views · 2026-07-30
自动驾驶autonomous driving单目感知
SLAM3R: Real-time 3D Scene from Single Video
1:10

SLAM3R: Real-time 3D Scene from Single Video

@LearnWithSubhan · 8 views · 2026-07-25
SLAM3D重建开源