Skip to search formSkip to main contentSkip to account menu
The Mapillary Vistas Dataset for Semantic Understanding of Street Scenes
TLDR
The Mapillary Vistas Dataset is a novel, large-scale street-level image dataset containing 25000 high-resolution images annotated into 66 object categories with additional, instance-specific labels for 37 classes, aiming to significantly further the development of state-of-the-art methods for visual road-scene understanding.
View on IEEE
Deep Neural Decision Forests
TLDR
A novel approach that unifies classification trees with the representation learning functionality known from deep convolutional networks, by training them in an end-to-end manner by introducing a stochastic and differentiable decision tree model.
View on IEEE
Disentangling Monocular 3D Object Detection
TLDR
An approach for monocular 3D object detection from a single RGB image, which leverages a novel disentangling transformation for 2D and 3D detection losses and a novel, self-supervised confidence score for 3D bounding boxes is proposed.
View on IEEE
In-place Activated BatchNorm for Memory-Optimized Training of DNNs
TLDR
This work presents In-Place Activated Batch Normalization (INPLACE-ABN), a novel approach to drastically reduce the training memory footprint of modern deep neural networks in a computationally efficient way, hence avoiding invasive framework surgery while providing straightforward applicability for existing deep learning frameworks.
View on IEEE
Structured class-labels in random forests for semantic image labelling
TLDR
This work provides a way to incorporate structural information in the popular random forest framework for performing low-level, unary classification and provides two possibilities for integrating the structured output predictions into concise, semantic labellings.
View on IEEE
AutoDIAL: Automatic Domain Alignment Layers
TLDR
Opposite to previous works which define a priori in which layers adaptation should be performed, this method is able to automatically learn the degree of feature alignment required at different levels of the deep network.
View on IEEE
Seamless Scene Segmentation
TLDR
This work introduces a novel, CNN-based architecture that can be trained end-to-end to deliver seamless scene segmentation results by means of a panoptic output format, going beyond the simple combination of independently trained segmentation and detection models.
View on IEEE
Modeling the Background for Incremental Learning in Semantic Segmentation
TLDR
This work revisits classical incremental learning methods, and proposes a new distillation-based framework which explicitly accounts for a semantic distribution shift, and introduces a novel strategy to initialize classifier's parameters, thus preventing biased predictions toward the background class.
View on IEEE
A Game-Theoretic Approach to Hypergraph Clustering
  • S. R. Bulò, M. Pelillo
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine…
  • 7 December 2009
TLDR
It is proved that the problem of finding the equilibria of the clustering game is equivalent to locally optimizing a polynomial function over the standard simplex, and a discrete-time high-order replicator dynamics to perform this optimization, based on the Baum-Eagon inequality is provided.
View on IEEE
Boosting Domain Adaptation by Discovering Latent Domains
TLDR
A novel Convolutional Neural Network (CNN) architecture is introduced which automatically discovers latent domains in visual datasets and exploits this information to learn robust target classifiers and outperforms state-of-the-art multi-source DA methods by a large margin.
View on IEEE
...
...
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy, Terms of Service, and Dataset License