Princeton365: A Diverse Dataset with Accurate Camera Pose
By: Karhan Kayan , Stamatis Alexandropoulos , Rishabh Jain and more
Potential Business Impact:
Helps robots map new places more accurately.
We introduce Princeton365, a large-scale diverse dataset of 365 videos with accurate camera pose. Our dataset bridges the gap between accuracy and data diversity in current SLAM benchmarks by introducing a novel ground truth collection framework that leverages calibration boards and a 360-camera. We collect indoor, outdoor, and object scanning videos with synchronized monocular and stereo RGB video outputs as well as IMU. We further propose a new scene scale-aware evaluation metric for SLAM based on the the optical flow induced by the camera pose estimation error. In contrast to the current metrics, our new metric allows for comparison between the performance of SLAM methods across scenes as opposed to existing metrics such as Average Trajectory Error (ATE), allowing researchers to analyze the failure modes of their methods. We also propose a challenging Novel View Synthesis benchmark that covers cases not covered by current NVS benchmarks, such as fully non-Lambertian scenes with 360-degree camera trajectories. Please visit https://princeton365.cs.princeton.edu for the dataset, code, videos, and submission.
Similar Papers
Leader360V: The Large-scale, Real-world 360 Video Dataset for Multi-task Learning in Diverse Environment
CV and Pattern Recognition
Helps robots and cars understand everything around them.
AthletePose3D: A Benchmark Dataset for 3D Human Pose Estimation and Kinematic Validation in Athletic Movements
CV and Pattern Recognition
Helps computers track athletes' fast moves better.
Dynamic Camera Poses and Where to Find Them
CV and Pattern Recognition
Creates realistic videos by tracking camera movement.