no code implementations • 25 May 2023 • Daniel Köhler, Maurice Quach, Michael Ulrich, Frank Meinl, Bastian Bischoff, Holger Blume
The proposed multi-scale KPPillarsBEV architecture outperforms the baseline by 5. 37% and the previous state of the art by 2. 88% in Car AP4. 0 (average precision for a matching threshold of 4 meters) on the nuScenes validation set.
no code implementations • 15 Aug 2023 • Marius Lippke, Maurice Quach, Sascha Braun, Daniel Köhler, Michael Ulrich, Bastian Bischoff, Wei Yap Tan
This paper investigates sparse convolutional object detection networks, which combine powerful grid-based detection with low compute resources.
1 code implementation • 1 Jul 2021 • Dat Thanh Nguyen, Maurice Quach, Giuseppe Valenzise, Pierre Duhamel
This paper proposes a lossless point cloud (PC) geometry compression method that uses neural networks to estimate the probability distribution of voxel occupancy.
1 code implementation • 11 Feb 2020 • Maurice Quach, Giuseppe Valenzise, Frederic Dufaux
However, as this mapping process is lossy in nature, we propose several strategies to refine it so that attributes can be mapped to the 2D grid with minimal distortion.
1 code implementation • 25 Feb 2021 • Maurice Quach, Aladine Chetouani, Giuseppe Valenzise, Frederic Dufaux
In addition, we propose a novel truncated distance field voxel grid representation and find that it leads to sparser latent spaces and loss functions that are more correlated with perceived visual quality compared to a binary representation.
1 code implementation • 30 Nov 2020 • Dat Thanh Nguyen, Maurice Quach, Giuseppe Valenzise, Pierre Duhamel
On the one hand, octree representation can eliminate the sparsity in the point cloud.
2 code implementations • 20 Apr 2021 • Dat Thanh Nguyen, Maurice Quach, Giuseppe Valenzise, Pierre Duhamel
We propose a practical deep generative approach for lossless point cloud geometry compression, called MSVoxelDNN, and show that it significantly reduces the rate compared to the MPEG G-PCC codec.
2 code implementations • 20 Mar 2019 • Maurice Quach, Giuseppe Valenzise, Frederic Dufaux
Efficient point cloud compression is fundamental to enable the deployment of virtual and mixed reality applications, since the number of points to code can range in the order of millions.
2 code implementations • 16 Jun 2020 • Maurice Quach, Giuseppe Valenzise, Frederic Dufaux
Point clouds have been recognized as a crucial data structure for 3D content and are essential in a number of applications such as virtual and mixed reality, autonomous driving, cultural heritage, etc.