• Publications
  • Influence
Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation
TLDR
This work proposes an end-to-end convolutional neural network for variable-length multi-frame video interpolation, where the motion interpretation and occlusion reasoning are jointly modeled.
SPLATNet: Sparse Lattice Networks for Point Cloud Processing
TLDR
A network architecture for processing point clouds that directly operates on a collection of points represented as a sparse set of samples in a high-dimensional lattice that outperforms existing state-of-the-art techniques on 3D segmentation tasks.
Gated-SCNN: Gated Shape CNNs for Semantic Segmentation
TLDR
A new two-stream CNN architecture for semantic segmentation that explicitly wires shape information as a separate processing branch, i.e. shape stream, that processes information in parallel to the classical stream is proposed.
Competitive Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentation
TLDR
This work addresses the unsupervised learning of several interconnected problems in low-level vision: single view depth prediction, camera motion estimation, optical flow, and segmentation of a video into the static scene and moving regions with Competitive Collaboration, a framework that facilitates the coordinated training of multiple specialized neural networks to solve complex problems.
Video Propagation Networks
TLDR
A Video Propagation Network that processes video frames in an adaptive manner that combines two components, a temporal bilateral network for dense and video adaptive filtering, followed by a spatial network to refine features and increased flexibility.
Pixel-Adaptive Convolutional Neural Networks
TLDR
A pixel-adaptive convolution (PAC) operation, a simple yet effective modification of standard convolutions, in which the filter weights are multiplied with a spatially varying kernel that depends on learnable, local pixel features, is proposed.
Training Deep Networks with Synthetic Data: Bridging the Reality Gap by Domain Randomization
TLDR
This work presents a system for training deep neural networks for object detection using synthetic images that relies upon the technique of domain randomization, in which the parameters of the simulator are randomized in non-realistic ways to force the neural network to learn the essential features of the object of interest.
Superpixel Sampling Networks
TLDR
A new differentiable model for superpixel sampling that leverages deep networks for learning superpixel segmentation and is end-to-end trainable, which allows learning task-specific superpixels with flexible loss functions and has fast runtime.
Learning Superpixels with Segmentation-Aware Affinity Loss
TLDR
This work proposes a segmentation-aware affinity learning approach for superpixel segmentation with a new loss function that takes the segmentation error into account for affinity learning and develops the Pixel Affinity Net for affinity prediction.
SCOPS: Self-Supervised Co-Part Segmentation
TLDR
This work proposes a self-supervised deep learning approach for part segmentation, where several loss functions are devised that aids in predicting part segments that are geometrically concentrated, robust to object variations and are also semantically consistent across different object instances.
...
1
2
3
4
5
...