Related papers: A three dimensional object point process for detec…
We present an improved approach for 3D object detection in point cloud data based on the Frustum PointNet (F-PointNet). Compared to the original F-PointNet, our newly proposed method considers the point neighborhood when computing point…
The redshift of all cosmological sources drifts by a systematic velocity of order a few m/s over a century due to the deceleration of the Universe. The specific functional dependence of the predicted velocity shift on the source redshift…
The 2dF Galaxy Redshift Survey has obtained 135,000 redshifts for galaxies in two broad strips. Here we present the first results of a 3-dimensional search for galaxy clusters based on known 2-dimensional compilations. We derive new…
Cosmology inference of galaxy clustering at the field level with the EFT likelihood in principle allows for extracting all non-Gaussian information from quasi-linear scales, while robustly marginalizing over any astrophysical uncertainties.…
In this paper, we propose a machine vision algorithm for automatically detecting defects in patterned textures with the help of gradient space and its energy. Experiments on real fabric images with defects show that the proposed method can…
We consider the problem of reliably finding filaments in point clouds. Realistic data sets often have numerous filaments of various sizes and shapes. Statistical techniques exist for finding one (or a few) filaments but these methods do not…
We propose a novel point cloud based 3D organ segmentation pipeline utilizing deep Q-learning. In order to preserve shape properties, the learning process is guided using a statistical shape model. The trained agent directly predicts…
The 2 degree Field (2dF) galaxy redshift survey will involve obtaining approximately 2.5 x 10^5 spectra of objects previously identified as galaxy candidates on morphological grounds. Included in these spectra should be about ten…
The universe in large scales is structured as a network known as cosmic web. Filaments are one of the structural components of this web, which can be introduced as a novel probe to study the formation and evolution of structures and as a…
A grand challenge of the 21st century cosmology is to accurately estimate the cosmological parameters of our Universe. A major approach to estimating the cosmological parameters is to use the large-scale matter distribution of the Universe.…
Object recognition and grasping plays a key role in robotic systems, especially for the autonomous robots to implement object sorting tasks in a warehouse. In this paper, we present a global texture-shape 3D feature descriptor which can be…
Recently, directly detecting 3D objects from 3D point clouds has received increasing attention. To extract object representation from an irregular point cloud, existing methods usually take a point grouping step to assign the points to an…
Determining cosmological parameters with high precision, as well as resolving current tensions in their values derived from low and high redshift probes, is one of the main objectives of the new generation of cosmological surveys. The…
In this paper, we focus on exploring the fusion of images and point clouds for 3D object detection in view of the complementary nature of the two modalities, i.e., images possess more semantic information while point clouds specialize in…
The large-scale structure is a major source of cosmological information. However, next-generation photometric galaxy surveys will only provide a distorted view of cosmic structures due to large redshift uncertainties. To address the need…
Current 3D object detection methods are heavily influenced by 2D detectors. In order to leverage architectures in 2D detectors, they often convert 3D point clouds to regular grids (i.e., to voxel grids or to bird's eye view images), or rely…
In this work, we study 3D object detection from RGB-D data in both indoor and outdoor scenes. While previous methods focus on images or 3D voxels, often obscuring natural 3D patterns and invariances of 3D data, we directly operate on raw…
In this paper, we present a real-time 3D detection approach considering time-spatial feature map aggregation from different time steps of deep neural model inference (named feature map flow, FMF). Proposed approach improves the quality of…
We propose a new approach for measuring the mass profile and shape of groups and clusters of galaxies, which uses lensing magnification of distant background galaxies. The main advantage of lensing magnification is that, unlike lensing…
Despite containing about a half of the total matter in the Universe, at most wavelengths the filamentary structure of the cosmic web is difficult to observe. In this work, we use large unigrid cosmological simulations to investigate how the…