Search Results for author:
Found 35 papers, 18 papers with code
Portmanteau Vocabularies for Multi-Cue Image Representation
We describe a novel technique for feature combination in the bag-of-words model of image classification.
Sparse Radial Sampling LBP for Writer Identification
In this paper we present the use of Sparse Radial Sampling Local Binary Patterns, a variant of Local Binary Patterns (LBP) for text-as-texture classification.
On-the-fly Network Pruning for Object Detection
Object detection with deep neural networks is often performed by passing a few thousand candidate bounding boxes through a deep neural network for each image.
Scale Coding Bag of Deep Features for Human Attribute and Action Recognition
Most approaches to human attribute and action recognition in still images are based on image representation in which multi-scale local features are pooled across scale into a single, scale-invariant encoding.
Bandwidth limited object recognition in high resolution imagery
This paper proposes a novel method to optimize bandwidth usage for object detection in critical communication scenarios.
Improving Text Proposals for Scene Images with Fully Convolutional Networks
Text Proposals have emerged as a class-dependent version of object proposals - efficient approaches to reduce the search space of possible text object locations in an image.
Visual attention models for scene text recognition
A set of feature vectors are derived from an intermediate convolutional layer corresponding to different areas of the image.
RankIQA: Learning from Rankings for No-reference Image Quality Assessment
Furthermore, on the LIVE benchmark we show that our approach is superior to existing NR-IQA techniques and that we even outperform the state-of-the-art in full-reference IQA (FR-IQA) methods without having to resort to high-quality reference images to infer IQA.
Review on Computer Vision Techniques in Emergency Situation
The number of emergencies where computer vision tools has been considered or used is very wide, and there is a great overlap across related emergency research.
Domain-adaptive deep network compression
We show that domain transfer leads to large shifts in network activations and that it is desirable to take this into account when compressing.
Rotate your Networks: Better Weight Consolidation and Less Catastrophic Forgetting
In this paper we propose an approach to avoiding catastrophic forgetting in sequential task learning scenarios.
Leveraging Unlabeled Data for Crowd Counting by Learning to Rank
We propose a novel crowd counting approach that leverages abundantly available unlabeled crowd imagery in a learning-to-rank framework.
Ranked #18 on
Crowd Counting
on UCF CC 50
Soft-PHOC Descriptor for End-to-End Word Spotting in Egocentric Scene Images
In this paper we propose a technique to create and exploit an intermediate representation of images based on text attributes which are character probability maps.
Exploiting Unlabeled Data in CNNs by Self-supervised Learning to Rank
Our results show that networks trained to regress to the ground truth targets for labeled data and to simultaneously learn to rank unlabeled data obtain significantly better, state-of-the-art results for both IQA and crowd counting.
Visual Question Answering for Cultural Heritage
This will turn the classic audio guide into a smart personal instructor with which the visitor can interact by asking for explanations focused on specific interests.
Generative Feature Replay For Class-Incremental Learning
To prevent forgetting, we combine generative feature replay in the classifier with feature distillation in the feature extractor.
RATT: Recurrent Attention to Transient Tasks for Continual Image Captioning
We call our method Recurrent Attention to Transient Tasks (RATT), and also show how to adapt continual learning approaches based on weight egularization and knowledge distillation to recurrent continual learning problems.
Class-incremental learning: survey and performance evaluation on image classification
For future learning systems, incremental learning is desirable because it allows for: efficient resource usage by eliminating the need to retrain from scratch at the arrival of new data; reduced memory usage by preventing or limiting the amount of data required to be stored -- also important when privacy limitations are imposed; and learning that more closely resembles human learning.
DeepCrawl: Deep Reinforcement Learning for Turn-based Strategy Games
In this paper we introduce DeepCrawl, a fully-playable Roguelike prototype for iOS and Android in which all agents are controlled by policy networks trained using Deep Reinforcement Learning (DRL).
Demonstration-efficient Inverse Reinforcement Learning in Procedurally Generated Environments
We propose a technique based on Adversarial Inverse Reinforcement Learning which can significantly decrease the need for expert demonstrations in PCG games.
Deep Policy Networks for NPC Behaviors that Adapt to Changing Design Parameters in Roguelike Games
Recent advances in Deep Reinforcement Learning (DRL) have largely focused on improving the performance of agents with the aim of replacing humans in known and well-defined environments.
Robust pedestrian detection in thermal imagery using synthesized images
Experimental results demonstrate the effectiveness of our approach: using less than 50\% of available real thermal training data, and relying on synthesized data generated by our model in the domain adaptation phase, our detector achieves state-of-the-art results on the KAIST Multispectral Pedestrian Detection Benchmark; even if more real thermal data is available adding GAN generated images to the training data results in improved performance, thus showing that these images act as an effective form of data augmentation.
Policy Fusion for Adaptive and Customizable Reinforcement Learning Agents
To this end, we propose four different policy fusion methods for combining pre-trained policies.
Continually Learning Self-Supervised Representations with Projected Functional Regularization
Recent self-supervised learning methods are able to learn high-quality image representations and are closing the gap with supervised approaches.
Planckian Jitter: countering the color-crippling effects of color jitter on self-supervised training
The data augmentations used are of crucial importance to the quality of learned feature representations.
CCPT: Automatic Gameplay Testing and Validation with Curiosity-Conditioned Proximal Trajectories
This paper proposes a novel deep reinforcement learning algorithm to perform automatic analysis and detection of gameplay issues in complex 3D navigation environments.
Towards Informed Design and Validation Assistance in Computer Games Using Imitation Learning
In games, as in and many other domains, design validation and testing is a huge challenge as systems are growing in size and manual testing is becoming infeasible.
Long-Tailed Class Incremental Learning
However, conventional CIL methods consider a balanced distribution for each new task, which ignores the prevalence of long-tailed distributions in the real world.
Exemplar-free Continual Learning of Vision Transformers via Gated Class-Attention and Cascaded Feature Drift Compensation
Secondly, we propose a new method of feature drift compensation that accommodates feature drift in the backbone when learning new tasks.
Task-Adaptive Saliency Guidance for Exemplar-free Class Incremental Learning
EFCIL is of interest because it mitigates concerns about privacy and long-term storage of data, while at the same time alleviating the problem of catastrophic forgetting in incremental learning.
Masked Autoencoders are Efficient Class Incremental Learners
Moreover, MAEs can reliably reconstruct original input images from randomly selected patches, which we use to store exemplars from past tasks more efficiently for CIL.
Class Incremental Learning with Pre-trained Vision-Language Models
We also propose a method for parameter retention in the adapter layers that uses a measure of parameter importance to better maintain stability and plasticity during incremental learning.
Small Dataset, Big Gains: Enhancing Reinforcement Learning by Offline Pre-Training with Model Based Augmentation
Offline reinforcement learning leverages pre-collected datasets of transitions to train policies.
Elastic Feature Consolidation for Cold Start Exemplar-free Incremental Learning
Experimental results on CIFAR-100, Tiny-ImageNet, ImageNet-Subset and ImageNet-1K demonstrate that Elastic Feature Consolidation is better able to learn new tasks by maintaining model plasticity and significantly outperform the state-of-the-art.
Task-conditioned Domain Adaptation for Pedestrian Detection in Thermal Imagery
Despite its broad application and interest, it remains a challenging problem in part due to the vast range of conditions under which it must be robust.
