• Publications
  • Influence
Multiscale conditional random fields for image labeling
TLDR
We propose an approach to include contextual features for labeling images, in which each pixel is assigned to one of a finite set of labels. Expand
Pose-Aware Multi-Level Feature Network for Human Object Interaction Detection
TLDR
We propose a multi-level relation detection strategy that utilizes human pose cues to capture global spatial configurations of relations and as an attention mechanism to dynamically zoom into relevant regions at human part level. Expand
Discrete-Continuous Depth Estimation from a Single Image
TLDR
We formulate monocular depth estimation as a discrete-continuous optimization problem, where the continuous variables encode the depth of the superpixels in the input image, and the discrete ones represent relationships between neighboring super pixels. Expand
SentiCap: Generating Image Descriptions with Sentiments
TLDR
We propose a novel switching recurrent neural network with word-level regularization, which is able to produce emotional image captions using only 2000+ training sentences containing sentiments. Expand
Robust Face Alignment Under Occlusion via Regional Predictive Power Estimation
TLDR
Face alignment has been well studied in recent years, however, when a face alignment model is applied on images with heavy partial occlusion, the performance deteriorates significantly. Expand
Dynamic Context Correspondence Network for Semantic Alignment
TLDR
In this paper, we aim to incorporate global semantic context in a flexible manner to overcome the limitations of prior work that relies on local semantic representations. Expand
Learning and Incorporating Top-Down Cues in Image Segmentation
TLDR
In this paper, we propose an approach to utilizing category-based information in segmentation, through a formulation as an image labelling problem. Expand
Superpixel Graph Label Transfer with Learned Distance Metric
TLDR
We present an algorithm for rapidly finding good matches by adaptively constructing a graph where nodes represent superpixels and edges represent matches. Expand
Geometry-Aware Deep Network for Single-Image Novel View Synthesis
TLDR
This paper tackles the problem of novel view synthesis from a single image. Expand
Part-aware Prototype Network for Few-shot Semantic Segmentation
TLDR
We propose a novel few-shot semantic segmentation framework based on the prototype representation, capable of capturing diverse and fine-grained object features. Expand
...
1
2
3
4
5
...