A High-speed and Computational Cost-effective 3D Recognition Method With 2D-Edge-based 4-points Congruent Set (4PCS)
With the increasing application of robots in industrial manufacturing, the 3D vision perception is becoming more and more important.
However, many researchers prefer to focus on the accuracy of 3D recognition rather than the speed.
Faster speed can accelerate industrial production a lot.
Meanwhile, the cost is also an important factor in reality.
Since these, we propose the 2D-Edge-based 4-points congruent set (4PCS) algorithm.
This algorithm utilizes the affine invariance and the projection invariant to find the correspondence between 3D model and 2D depth image, resulting in a significant increase in speed and keeping the accuracy.
Its necessary hardware are just a consumer-level CPU and a consumer-level depth camera.
On the Sileana dataset, the average recognition time is 1 ms per image.
For the experiment in reality, our algorithm can recognize a flying object precisely at 3 kHz.
Also, we achieve non-stop object picking using the proposed method.
Movie
References
Xiaohang Shi, Qitong Guo, Kenichi Murakami and Yuji Yamakawa: A High-speed and Computational Cost-effective 3D Recognition Method With 2D-Edges-based 4-points Congruent Set Algorithm, IEEE Access (2024)