Deep 6-DOF Tracking

Abstract

We present a temporal 6-DOF tracking method which leverages deep learning to achieve state-of-the-art performance on challenging datasets of real world capture. Our method is both more accurate and more robust to occlusions than the existing best performing approaches while maintaining real-time performance. To assess its efficacy, we evaluate our approach on several challenging RGBD sequences of real objects in a variety of conditions. Notably, we systematically evaluate robustness to occlusions through a series of sequences where the object to be tracked is increasingly occluded. Finally, our approach is purely data-driven and does not require any hand-designed features: robust tracking is automatically learned from data.

9 pages, 9 figures, ISMAR 2017, TVCG special edition Website: http://vision.gel.ulaval.ca/~jflalonde/projects/deepTracking/index.html

References in corpus (5)

• Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
• Deep Image Homography Estimation
• First Step toward Model-Free, Anonymous Object Tracking with Recurrent Neural Networks
• Siamese Regression Networks with Efficient mid-level Feature Extraction for 3D Object Pose Estimation
• PoseAgent: Budget-Constrained 6D Object Pose Estimation via Reinforcement Learning

Cited by in corpus (9)

• DeepIM: Deep Iterative Matching for 6D Pose Estimation
• SRT3D: A Sparse Region-Based 3D Object Tracking Approach for the Real World
• I Like to Move It: 6D Pose Estimation as an Action Decision Process
• se(3)-TrackNet: Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains
• Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains
• DI-Fusion: Online Implicit 3D Reconstruction with Deep Priors
• Physics-Based Rigid Body Object Tracking and Friction Filtering From RGB-D Videos
• A Wide-area, Low-latency, and Power-efficient 6-DoF Pose Tracking System for Rigid Objects
• Depth Camera Based Particle Filter for Robotic Osteotomy Navigation