Skip to search formSkip to main contentSkip to account menu

Image Inpainting for Irregular Holes Using Partial Convolutions

This work proposes the use of partial convolutions, where the convolution is masked and renormalized to be conditioned on only valid pixels, and outperforms other methods for irregular masks.
View PDF on arXiv (opens in a new tab)

Improving Semantic Segmentation via Video Propagation and Label Relaxation

This paper presents a video prediction-based methodology to scale up training sets by synthesizing new training samples in order to improve the accuracy of semantic segmentation networks, and introduces a novel boundary label relaxation technique that makes training robust to annotation noise and propagation artifacts along object boundaries.
View on IEEE (opens in a new tab)

Graphical Contrastive Losses for Scene Graph Parsing

A set of contrastive loss formulations are proposed that specifically target these types of errors within the scene graph parsing problem, collectively termed the Graphical Contrastive Losses, and show improved results over the best previous methods on the Visual Genome and Visual Relationship Detection datasets.
View on IEEE (opens in a new tab)

Where to Look: Focus Regions for Visual Question Answering

A method that learns to answer visual questions by selecting image regions relevant to the text-based query that exhibits significant improvements in answering questions such as "what color", where it is necessary to evaluate a specific location, and "what room," where it selectively identifies informative image regions.
View on IEEE (opens in a new tab)

SDC-Net: Video Prediction Using Spatially-Displaced Convolution

SDC module for video frame prediction with spatially-displaced convolution inherits the merits of both vector-based and kernel-based approaches, while ameliorating their respective disadvantages.
View PDF on arXiv (opens in a new tab)

Learning Collections of Part Models for Object Recognition

The detection system is competitive with the best-existing systems, outperforming other HOG-based detectors on the more deformable categories and evaluating the part detectors' ability to discriminate and localize annotated key points on PASCAL VOC 2010.
View on IEEE (opens in a new tab)

Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis

The mean opinion scores (MOS) show that Flowtron matches state-of-the-art TTS models in terms of speech quality, and results on control of speech variation, interpolation between samples and style transfer between speakers seen and unseen during training are provided.
View PDF on arXiv (opens in a new tab)

One TTS Alignment to Rule Them All

This paper leverages the alignment mechanism proposed in RAD-TTS and improves alignment convergence speed, simplifies the training pipeline by eliminating need for external aligners, enhances robustness to errors on long utterances and improves the perceived speech synthesis quality, as judged by human evaluators.
View on IEEE (opens in a new tab)

Graphical Contrastive Losses for Scene Graph Generation

A set of contrastive loss formulations are proposed that specifically target these types of errors within the scene graph generation problem, collectively termed the Graphical Contrastive Losses.
View PDF on arXiv (opens in a new tab)

Partial Convolution based Padding

This paper presents a simple yet effective padding scheme that can be used as a drop-in module for existing convolutional neural networks and demonstrates that the proposed padding scheme consistently outperforms standard zero padding with better accuracy.
View PDF on arXiv (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)