Presenting Pose Estimation work at BMVC and ECCV meetings:
Chaitanya Mitash, who is working on the problem of 6D Pose Estimation, presented his latest paper on “Robust 6D Object Pose Estimation with Stochastic Congruent Sets” at the British Machine Vision Conference (BMVC) 2018 in Newcastle, England, UK on September 5th.
This work proposes a novel stochastic optimization process that treats the segmentation output of CNNs as a confidence probability. The algorithm, called Stochastic Congruent Sets (StoCS), samples point-sets on the point cloud according to the soft segmentation distribution and so as to agree with the object’s known geometry. The point-sets are then matched to congruent sets on the 3D object model to generate pose estimates. StoCS is shown to be robust on an APC dataset, despite the fact the CNN is trained only on synthetic data. In the YCB dataset, StoCS outperforms a recent network for 6D pose estimation and alternative point-set matching techniques. Below is the corresponding poster:
Kostas Bekris presented an overview of Chaitanya’s work on Robust Pose Estimation, which has taken place in collaboration with Prof. Abdeslam Boularias, during an ECCV 2018 workshop on Recovering 6D Object Pose in Munich, Germany on September 9th. You can find below an image from the presentation.
If you are interested in the presentation material, please contact Kostas Bekris.