Real-time RGBD Object Recognition

Visit sites.google.c... for more details. Weakly-Supervised Deep Learning for RGBD Object Recognition. This research addresses the problem of RGBD object recognition in real-world applications, where large amounts of annotated training data are typically unavailable.
The dataset, paper, and source code can be found in: sites.google.c...
Paper link:
ieeexplore.iee...

[@ingenganu]

Good job Li... U know about some approach implementation with no need of firstly segmentation and removing of large planes? A great goal could be object pose recognition and tracking in real world random scenarios, where object could be moving or standing over random surfaces, even could be flying, with no large planes in the scene...

[@lisun8087]

Yes, the end-to-end method can work without plane removal, but much more examples are required.

[@zariframzan7777]

Awesome works, i can't access the code......can you share the code again