Searcharxiv⌕ Search

arXiv subjects

Li Yi

Publications and source records attributed to Li Yi.

At least 145 records · Page 8Linked to original sources
arXiv2018

GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloud

We introduce a novel 3D object proposal approach named Generative Shape Proposal Network (GSPN) for instance segmentation in point cloud data. Instead of treating object proposal as a direct bounding box regression problem, we take an analysis-by-synthesis strategy and generate proposals by reconstructing shapes from noisy observations in a scene. We incorporate GSPN into a novel 3D instance segmentation framework named Region-based PointNet (R-PointNet) which allows flexible proposal refinement and instance segmentation generation. We achieve state-of-the-art performance on several 3D instance segmentation tasks. The success of GSPN largely comes from its emphasis on geometric understandings during object proposal, which greatly reducing proposals with low objectness.

cs.CV↗
arXiv2018

PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understanding

We present PartNet: a consistent, large-scale dataset of 3D objects annotated with fine-grained, instance-level, and hierarchical 3D part information. Our dataset consists of 573,585 part instances over 26,671 3D models covering 24 object categories. This dataset enables and serves as a catalyst for many tasks such as shape analysis, dynamic 3D scene modeling and simulation, affordance analysis, and others. Using our dataset, we establish three benchmarking tasks for evaluating 3D part recognition: fine-grained semantic segmentation, hierarchical semantic segmentation, and instance segmentation. We benchmark four state-of-the-art 3D deep learning algorithms for fine-grained semantic segmentation and three baseline methods for hierarchical semantic segmentation. We also propose a novel method for part instance segmentation and demonstrate its superior performance over existing methods.

cs.CV↗
arXiv2018

Deep Part Induction from Articulated Object Pairs

Object functionality is often expressed through part articulation -- as when the two rigid parts of a scissor pivot against each other to perform the cutting function. Such articulations are often similar across objects within the same functional category. In this paper, we explore how the observation of different articulation states provides evidence for part structure and motion of 3D objects. Our method takes as input a pair of unsegmented shapes representing two different articulation states of two functionally related objects, and induces their common parts along with their underlying rigid motion. This is a challenging setting, as we assume no prior shape structure, no prior shape category information, no consistent shape orientation, the articulation states may belong to objects of different geometry, plus we allow inputs to be noisy and partial scans, or point clouds lifted from RGB images. Our method learns a neural network architecture with three modules that respectively propose correspondences, estimate 3D deformation flows, and perform segmentation. To achieve optimal performance, our architecture alternates between correspondence, deformation flow, and segmentation prediction iteratively in an ICP-like fashion. Our results demonstrate that our method significantly outperforms state-of-the-art techniques in the task of discovering articulated parts of objects. In addition, our part induction is object-class agnostic and successfully generalizes to new and unseen objects.

cs.CV↗
arXiv2017

Large-Scale 3D Shape Reconstruction and Segmentation from ShapeNet Core55

We introduce a large-scale 3D shape understanding benchmark using data and annotation from ShapeNet 3D object database. The benchmark consists of two tasks: part-level segmentation of 3D shapes and 3D reconstruction from single view images. Ten teams have participated in the challenge and the best performing teams have outperformed state-of-the-art approaches on both tasks. A few novel deep learning architectures have been proposed on various 3D representations on both tasks. We report the techniques used by each team and the corresponding performances. In addition, we summarize the major discoveries from the reported results and possible trends for the future work in the field.

cs.CV↗
arXiv2017

PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

Few prior works study deep learning on point sets. PointNet by Qi et al. is a pioneer in this direction. However, by design PointNet does not capture local structures induced by the metric space points live in, limiting its ability to recognize fine-grained patterns and generalizability to complex scenes. In this work, we introduce a hierarchical neural network that applies PointNet recursively on a nested partitioning of the input point set. By exploiting metric space distances, our network is able to learn local features with increasing contextual scales. With further observation that point sets are usually sampled with varying densities, which results in greatly decreased performance for networks trained on uniform densities, we propose novel set learning layers to adaptively combine features from multiple scales. Experiments show that our network called PointNet++ is able to learn deep point set features efficiently and robustly. In particular, results significantly better than state-of-the-art have been obtained on challenging benchmarks of 3D point clouds.

cs.CV↗
arXiv2017

Learning Hierarchical Shape Segmentation and Labeling from Online Repositories

We propose a method for converting geometric shapes into hierarchically segmented parts with part labels. Our key idea is to train category-specific models from the scene graphs and part names that accompany 3D shapes in public repositories. These freely-available annotations represent an enormous, untapped source of information on geometry. However, because the models and corresponding scene graphs are created by a wide range of modelers with different levels of expertise, modeling tools, and objectives, these models have very inconsistent segmentations and hierarchies with sparse and noisy textual tags. Our method involves two analysis steps. First, we perform a joint optimization to simultaneously cluster and label parts in the database while also inferring a canonical tag dictionary and part hierarchy. We then use this labeled data to train a method for hierarchical segmentation and labeling of new 3D shapes. We demonstrate that our method can mine complex information, detecting hierarchies in man-made objects and their constituent parts, obtaining finer scale details than existing alternatives. We also show that, by performing domain transfer using a few supervised examples, our technique outperforms fully-supervised techniques that require hundreds of manually-labeled models.

cs.GR↗
arXiv2016

Beyond Holistic Object Recognition: Enriching Image Understanding with Part States

Important high-level vision tasks such as human-object interaction, image captioning and robotic manipulation require rich semantic descriptions of objects at part level. Based upon previous work on part localization, in this paper, we address the problem of inferring rich semantics imparted by an object part in still images. We propose to tokenize the semantic space as a discrete set of part states. Our modeling of part state is spatially localized, therefore, we formulate the part state inference problem as a pixel-wise annotation problem. An iterative part-state inference neural network is specifically designed for this task, which is efficient in time and accurate in performance. Extensive experiments demonstrate that the proposed method can effectively predict the semantic states of parts and simultaneously correct localization errors, thus benefiting a few visual understanding applications. The other contribution of this paper is our part state dataset which contains rich part-level semantic annotations.

cs.CV↗
arXiv2016

SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation

In this paper, we study the problem of semantic annotation on 3D models that are represented as shape graphs. A functional view is taken to represent localized information on graphs, so that annotations such as part segment or keypoint are nothing but 0-1 indicator vertex functions. Compared with images that are 2D grids, shape graphs are irregular and non-isomorphic data structures. To enable the prediction of vertex functions on them by convolutional neural networks, we resort to spectral CNN method that enables weight sharing by parameterizing kernels in the spectral domain spanned by graph laplacian eigenbases. Under this setting, our network, named SyncSpecCNN, strive to overcome two key challenges: how to share coefficients and conduct multi-scale analysis in different parts of the graph for a single shape, and how to share information across related but different shapes that may be represented by very different graphs. Towards these goals, we introduce a spectral parameterization of dilated convolutional kernels and a spectral transformer network. Experimentally we tested our SyncSpecCNN on various tasks, including 3D shape part segmentation and 3D keypoint prediction. State-of-the-art performance has been achieved on all benchmark datasets.

cs.CV↗
arXiv2016

Dialogue Session Segmentation by Embedding-Enhanced TextTiling

In human-computer conversation systems, the context of a user-issued utterance is particularly important because it provides useful background information of the conversation. However, it is unwise to track all previous utterances in the current session as not all of them are equally important. In this paper, we address the problem of session segmentation. We propose an embedding-enhanced TextTiling approach, inspired by the observation that conversation utterances are highly noisy, and that word embeddings provide a robust way of capturing semantics. Experimental results show that our approach achieves better performance than the TextTiling, MMD approaches.

cs.CL↗
arXiv2015

ShapeNet: An Information-Rich 3D Model Repository

We present ShapeNet: a richly-annotated, large-scale repository of shapes represented by 3D CAD models of objects. ShapeNet contains 3D models from a multitude of semantic categories and organizes them under the WordNet taxonomy. It is a collection of datasets providing many semantic annotations for each 3D model such as consistent rigid alignments, parts and bilateral symmetry planes, physical sizes, keywords, as well as other planned annotations. Annotations are made available through a public web-based interface to enable data visualization of object attributes, promote data-driven geometric analysis, and provide a large-scale quantitative benchmark for research in computer graphics and vision. At the time of this technical report, ShapeNet has indexed more than 3,000,000 models, 220,000 models out of which are classified into 3,135 categories (WordNet synsets). In this report we describe the ShapeNet effort as a whole, provide details for all currently available datasets, and summarize future plans.

cs.GR↗
arXiv2015

STAR Overview of Hard Probe Observables

Parton energy loss, quarkonium sequential melting and particle production from electromagnetic interactions are tools to study Quark Gluon Plasma properties. The STAR detector, with large acceptance at mid-rapidity, excellent particle identification and wide transverse momentum coverage, is able to study these probes in details. In Hard Probes 2015, the STAR collaboration reported measurements of reconstructed jets, heavy-flavor physics, di-lepton production and the performance of new detectors in seven presentations and one poster. Given the rich results from STAR, this overview report will focus on a few selected results on jets and $Υ$ measurements in Au+Au collisions at $\sqrt{s_{\rm NN}}=200\,$GeV, $J/ψ$ production in $p$+$p$ collisions at $\sqrt{s_{\rm NN}}=500\,$GeV, and the di-electron spectrum in the low mass region from the Beam Energy Scan - Phase I.

nucl-ex↗
arXiv2014

3D-Assisted Image Feature Synthesis for Novel Views of an Object

Comparing two images in a view-invariant way has been a challenging problem in computer vision for a long time, as visual features are not stable under large view point changes. In this paper, given a single input image of an object, we synthesize new features for other views of the same object. To accomplish this, we introduce an aligned set of 3D models in the same class as the input object image. Each 3D model is represented by a set of views, and we study the correlation of image patches between different views, seeking what we call surrogates --- patches in one view whose feature content predicts well the features of a patch in another view. In particular, for each patch in the novel desired view, we seek surrogates from the observed view of the given image. For a given surrogate, we predict that surrogate using linear combination of the corresponding patches of the 3D model views, learn the coefficients, and then transfer these coefficients on a per patch basis to synthesize the features of the patch in the novel view. In this way we can create feature sets for all views of the latent object, providing us a multi-view representation of the object. View-invariant object comparisons are achieved simply by computing the $L^2$ distances between the features of corresponding views. We provide theoretical and empirical analysis of the feature synthesis process, and evaluate the proposed view-agnostic distance (VAD) in fine-grained image retrieval (100 object classes) and classification tasks. Experimental results show that our synthesized features do enable view-independent comparison between images and perform significantly better than traditional image features in this respect.

cs.CV↗
arXiv2013

Possibility to disentangle anisotropic flow, flow fluctuation, and nonflow assuming Gaussian fluctuations

We suggest the possibility to disentangle anisotropic flow, flow fluctuation, and nonflow using two-, four-, and six-particle azimuthal moments assuming Gaussian fluctuations. We show that such disentanglement is possible when the flow fluctuations are large, comparable to the average flow magnitude. When fluctuations are small, the disentanglement becomes difficult. We verify our results with a toy-model Monte Carlo simulation.

nucl-ex↗
arXiv2013

A Bando-Mabuchi Uniqueness Theorem

In this paper we prove a uniqueness theorem on generalized Kahler-Einstein metrics on Fano manifolds. Our result generalize the one shown by Berndtsson using the convexity properties of Bergman kernels. The same technics as well as that of the regularization of closed positive currents and an estimate due to Demailly-Kollar will be used in our proof.

math.CV↗
arXiv2012

Isolation of Flow and Nonflow by Two- and Multi-Particle Cumulant Measurements of $v_n$ in $\sqrt{s_{NN}} = 200$ GeV Au+Au Collisions by STAR

We apply a data-driven method to STAR Au+Au collisions at $\sqrt{s_{NN}}=200$ GeV to isolate $Δη$-dependent and $Δη$-independent correlations by using two- and four-particle Q-cumulant $v_{n}$ measurements. The $Δη$-independent part, dominated by flow, is found to be $η$-independent within the STAR TPC of $\pm1$ unit of pseudo-rapidity. The $Δη$-dependent part may be associated to nonflow, and is responsible for the $Δη$ drop in the measured two-particle $v_{2}$ cumulant. We combine our result to four- and six-particle cumulants to gain further insights on the nature of flow fluctuations.

nucl-ex↗
arXiv2012

Decomposition of flow and nonflow in relativistic heavy-ion collisions

We propose a method to separate Δη-dependent and Δη-independent azimuthal correlations using two- and four-particle cumulants between pseudo-rapidity (η) bins in symmetric heavy-ion collisions. The Δη-independent correlation may be dominated by harmonic flows, a global correlation to the common collision geometry. The Δη-dependent correlation can be identified as nonflow, particle correlations unrelated to the common geometry. Our method exploits the ηsymmetry of the average harmonic flows and is "data-driven." We use the AMPT and HIJING event generators to illustrate our method. We discuss the decomposed Δη-independent and Δη-dependent correlations regarding flow and nonflow in the models.

nucl-ex↗
arXiv2011

Statistical Origin of Constituent-Quark Scaling in the QGP hadronization

Nonextensive statistics in a Blast-Wave model (TBW) is implemented to describe the identified hadron production in relativistic p+p and nucleus-nucleus collisions. Incorporating the core and corona components within the TBW formalism allows us to describe simultaneously some of the major observations in hadronic observables at the Relativistic Heavy-Ion Collider (RHIC): the Number of Constituent Quark Scaling (NCQ), the large radial and elliptic flow, the effect of gluon saturation and the suppression of hadron production at high transverse momentum (pT) due to jet quenching. In this formalism, the NCQ scaling at RHIC appears as a consequence of non-equilibrium process. Our study also provides concise reference distributions with a least chi2 fit of the available experimental data for future experiments and models.

nucl-ex↗