no code implementations • 2 Nov 2021 • Enxu Li, Ryan Razani, YiXuan Xu, Bingbing Liu
A fast and accurate panoptic segmentation system for LiDAR point clouds is crucial for autonomous driving vehicles to understand the surrounding objects and scenes.
no code implementations • 31 Aug 2021 • Enxu Li, Ryan Razani, YiXuan Xu, Liu Bingbing
Thus, we propose to use a novel centroid-aware repel loss as an additional term to effectively supervise the network to differentiate each object cluster with its neighbours.
no code implementations • ICCV 2021 • Ryan Razani, Ran Cheng, Enxu Li, Ehsan Taghavi, Yuan Ren, Liu Bingbing
GP-S3Net is a proposal-free approach in which no object proposals are needed to identify the objects in contrast to conventional two-stage panoptic systems, where a detection network is incorporated for capturing instance information.
no code implementations • 16 Mar 2021 • Ryan Razani, Ran Cheng, Ehsan Taghavi, Liu Bingbing
Autonomous driving vehicles and robotic systems rely on accurate perception of their surroundings.
no code implementations • 15 Mar 2021 • Ran Cheng, Ryan Razani, Yuan Ren, Liu Bingbing
In literature, several approaches are introduced to attempt LiDAR semantic segmentation task, such as projection-based (range-view or birds-eye-view), and voxel-based approaches.
no code implementations • 8 Feb 2021 • Ran Cheng, Ryan Razani, Ehsan Taghavi, Enxu Li, Bingbing Liu
Autonomous robotic systems and self driving cars rely on accurate perception of their surroundings as the safety of the passengers and pedestrians is the top priority.
Ranked #3 on 3D Semantic Segmentation on nuScenes
no code implementations • 24 Aug 2020 • Martin Gerdzhev, Ryan Razani, Ehsan Taghavi, Bingbing Liu
Semantic segmentation of point clouds is a key component of scene understanding for robotics and autonomous driving.
Ranked #15 on 3D Semantic Segmentation on SemanticKITTI
no code implementations • 26 Sep 2019 • Ryan Razani, Grégoire Morin, Vahid Partovi Nia, Eyyüb Sari
Ternary quantization provides a more flexible model and outperforms binary quantization in terms of accuracy, however doubles the memory footprint and increases the computational cost.
no code implementations • 25 Sep 2019 • Gregoire Morin, Ryan Razani, Vahid Partovi Nia, Eyyub Sari
Low bit quantization such as binary and ternary quantization is a common approach to alleviate this resource requirements.
no code implementations • CVPR 2020 • Ehsan Nezhadarya, Ehsan Taghavi, Ryan Razani, Bingbing Liu, Jun Luo
While several convolution-like operators have recently been proposed for extracting features out of point clouds, down-sampling an unordered point cloud in a deep neural network has not been rigorously studied.