TL;DR
A method to train a 3D object detector from a single RGB camera without domain-specific human annotations, making orders of magnitude more data available for training and shows that combining pseudo-labels from multiple datasets can achieve comparable accuracy to using human labels from a single dataset.
Three-dimensional object detection for autonomous driving is dominated by detectors trained on large corpora of human-annotated 3D boxes, which can be solved training-free and open-vocabulary.
Traffic video often lacks camera intrinsics, installation height, and surveyed scale. We study coarse metric ranging from a single RGB frame under this zero-user-calibration setting. The proposed pipeline combines GeoCalib, Depth Anything V2 Metric, robust bottom-contact sampling, and 3-D backprojection, withholding es...
Multi-camera 3D perception systems for warehouse scenes are trained largely on synthetic data and evaluated on physically captured environments. The resulting synthetic-to-real gap, which corrupts ground-plane localization and cross-camera identity association, is usually treated as one deficiency for a single domain-a...
Map-Det3D is an online multi-view 3D object detection model that brings detection directly into a 3D space reconstructed from RGB, suggesting that training reconstruction priors for detection is a practical route to stable metric 3D detection from monocular video.
Results from an adaptation of DINOv3 to the instance-level fusion pipeline of IIFNet3D are interpreted as evidence for a controlled transfer recipe rather than as a causal estimate of the individual contributions of VAID or depth filtering.
Extensive experimental results show that training on the automatically generated optical flow outperforms methods that are trained on rigid synthetic data using FlowNet-S, LiteFlowNet, PWC-Net, PWC-Net, and RAFT.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
