/ussd-6d

Primary LanguagePython

Unsupervised Single Shot 6D Object Pose Prediction

Developed by Jiatian Wu and Youjie Xia.

References

[1] Tekin et al., "Real-Time Seamless Single Shot 6D Object Pose Prediction", arXiv 2017.

Dataset

Refer to dataset_info.md or SIXD Challenge 2017: format of datasets. And we preprocess the original data and save the 2D projections of the 3D bounding box vertices and centroids, and ground truth grid in the '.npy' format (in the './data' folder).