Improving Self-Supervised Learning of Transparent Category Poses With Language Guidance and Implicit Physical Constraints

Keyphrases

• Physical Constraints 100%
• Self-supervision 100%
• Self-supervised Learning 100%
• Language Guidance 100%
• Background Change 40%
• Robotic Applications 20%
• Superior Performance 20%
• Object Pose Estimation 20%
• Object Segmentation 20%
• Sim-to-real 20%
• Domain Gap 20%
• Recent Trends 20%
• Depth Measurement 20%
• Home Environment 20%
• Transparent Object 20%
• Category-level Pose Estimation 20%
• 2D Objects 20%
• Normal Contact 20%
• Contact Constraints 20%
• Self-collected 20%
• Supervision Approaches 20%
• Differentiable Renderer 20%
• Similar Object 20%
• Normal Constraint 20%
• Application Trend 20%

Computer Science

• Self-Supervised Learning 100%
• Physical Constraint 100%
• Pose Estimation 66%
• Superior Performance 33%
• Object Pose 33%
• Robotics Application 33%
• Transparent Object 33%
• Depth Measurement 33%
• Home Environment 33%