Published
For the first time, an algorithm trained only on synthetic data is able to beat a state-of-the-art network trained on real images for object pose estimation on several objects of a standard benchmark. Learn more here: https://nvda.ws/2CvO2Jy
- Category
- Hardware
Sign in or sign up to post comments.
Be the first to comment
Up Next
-
02:17
Scan Objects into Realistic 3D Models with HP Z and NVIDIA GPUs
-
02:15
Research at NVIDIA: AI Reconstructs Photos with Realistic Results
-
01:26
NVIDIA DRIVE Labs: Tracking Objects With Surround Camera Vision
-
03:07
The Alvin Submarine Part 3: Humans vs. Robots and the Future of Deep-Sea Research – WIRED
-
16:57
Wallet Objects Concepts: Templates, Classes and Objects
-
01:33
Research at NVIDIA: The First Interactive AI Rendered Virtual World
-
01:39
Research at NVIDIA: Improving Landmark Localization with a New Deep Learning Architecture
-
01:10
Research at NVIDIA: GPU-Accelerated Robotic Simulation for Distributed Reinforcement Learning
-
18:06
Using advanced pre-trained Web ML models - Part 2: Using MoveNet for pose estimation in the browser
-
02:01
Accelerating AI for Biomedical Research with The Broad Institute and NVIDIA
-
01:20
NVIDIA Graphics Research
-
36:54
Gemini app: Canvas, Deep Research and personalization
-
02:40
Robotics Research at NVIDIA: Closing the Sim-to-Real Loop
-
00:58
NVIDIA Research Breakthroughs at CVPR 2025
-
01:34
Accelerating Deep Learning Research with NVIDIA DGX Station A100
-
02:27
Celebrating Robotics in Seattle: NVIDIA Opens New AI Research Lab
-
02:29
Research at NVIDIA: Structured Domain Randomization
-
02:40
Research at NVIDIA: New Core AI and Machine Learning Lab
-
01:55
Trail Blazing our Future with AI and NVIDIA Research
-
01:48
Berkeley AI Research (BAIR) Teaches Robot to Interact with Real World Objects
-
10:16
Archaeologists Found This Bizarre Object Deep Inside Mount Owen Cave System!
-
01:39
How NVIDIA Research is Teaching Robots to Interact with Humans
Loading...