• Publications
  • Influence
Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images
TLDR
An end-to-end deep learning architecture that produces a 3D shape in triangular mesh from a single color image by progressively deforming an ellipsoid, leveraging perceptual features extracted from the input image.
Combinatorial Optimization with Graph Convolutional Networks and Guided Tree Search
TLDR
Experimental results demonstrate that the presented approach substantially outperforms recent deep learning work, and performs on par with highly optimized state-of-the-art heuristic solvers for some NP-hard problems.
What Do Single-View 3D Reconstruction Networks Learn?
TLDR
This work sets up two alternative approaches that perform image classification and retrieval respectively and shows that encoder-decoder methods are statistically indistinguishable from these baselines, indicating that the current state of the art in single-view object reconstruction does not actually perform reconstruction but image classification.
Interactive Image Segmentation with Latent Diversity
TLDR
The proposed architecture couples two convolutional networks and is trained to synthesize a diverse set of plausible segmentations that conform to the user's input, which retains compatibility with existing interactive segmentation interfaces.
PointPWC-Net: Cost Volume on Point Clouds for (Self-)Supervised Scene Flow Estimation
TLDR
A novel end-to-end deep scene flow model, called PointPWC-Net, that directly processes 3D point cloud scenes with large motions in a coarse- to-fine fashion, and shows great generalization ability on the KITTI Scene Flow 2015 dataset, outperforming all previous methods.
Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness
TLDR
The proposed ATV consists of only a small number of planes with low memory and computation costs; yet, it efficiently partitions local depth ranges within learned small uncertainty intervals, which enables reconstruction with high completeness and accuracy in a coarse-to-fine fashion.
PointPWC-Net: A Coarse-to-Fine Network for Supervised and Self-Supervised Scene Flow Estimation on 3D Point Clouds
TLDR
This work proposes a novel end-to-end deep scene flow model, called PointPWC-Net, on 3D point clouds in a coarse- to-fine fashion, which shows great generalization ability on KITTI Scene Flow 2015 dataset, outperforming all previous methods.
Pixel2Mesh++: Multi-View 3D Mesh Generation via Deformation
TLDR
This model learns to predict series of deformations to improve a coarse shape iteratively and exhibits generalization capability across different semantic categories, number of input images, and quality of mesh initialization.
Simultaneous video defogging and stereo reconstruction
TLDR
A method to jointly estimate scene depth and recover the clear latent image from a foggy video sequence, which is optimized iteratively by introducing auxiliary variables in an MRF framework.
Perspective Motion Segmentation via Collaborative Clustering
TLDR
This paper first formulates the 3-D motion segmentation from two perspective views as a subspace clustering problem, utilizing the epipolar constraint of an image pair, and proposes an over-segment and merge approach, where the merging step is based on the property of the ell_1-norm of the mutual sparse representation of two over-Segmented groups.
...
...