Incremental Multi-Target Domain Adaptation for Object Detection with Efficient Domain Transfer (2021-04-13T00:00:00.000000Z)

TL;DR

This work proposes a new Incremental MTDA technique for object detection that can adapt a detector to multiple target domains, one at a time, without having to retain data of previously-learned target domains.

Authors

Eric Granger

9 papers

Marco Pedersoli

4 papers

J. Dolz

14 papers

Le Thanh Nguyen-Meidine

2 papers

M. Kiran

2 papers

Louis-Antoine Blais-Morin

2 papers

TL;DR

Authors

References65 items

Generative Adversarial Networks

Multi-Target Domain Adaptation with Collaborative Consistency Learning

Scale variance minimization for unsupervised domain adaptation in image segmentation

Uncertainty-Aware Unsupervised Domain Adaptation in Object Detection

Instance-Invariant Domain Adaptive Object Detection Via Progressive Disentanglement

Generative attention adversarial classification network for unsupervised domain adaptation

Incremental learning based multi-domain adaptation for object detection

Simplified unsupervised image translation for semantic segmentation adaptation

Unsupervised Multi-Target Domain Adaptation Through Knowledge Distillation

Joint Progressive Knowledge Distillation and Unsupervised Domain Adaptation

YOLOv4: Optimal Speed and Accuracy of Object Detection

Dark Experience for General Continual Learning: a Strong, Simple Baseline

Semantic Drift Compensation for Class-Incremental Learning

Self-Guided Adaptation: Progressive Representation Alignment for Domain Adaptive Object Detection

Harmonizing Transferability and Discriminability for Adapting Object Detectors

Incremental Few-Shot Object Detection

iFAN: Image-Instance Full Alignment Networks for Adaptive Object Detection

SpotNet: Self-Attention Multi-Task Network for Object Detection

Multi-task incremental learning for object detection

Deep conditional adaptation networks and label correlation transfer for unsupervised domain adaptation

Correlation-aware Adversarial Domain Adaptation and Generalization

EfficientDet: Scalable and Efficient Object Detection

Maintaining Discrimination and Fairness in Class Incremental Learning

Progressive Domain Adaptation for Object Detection

Multi-Adversarial Faster-RCNN for Unrestricted Object Detection

Blending-Target Domain Adaptation by Adversarial Meta-Adaptation Networks

Depth-Attentional Features for Single-Image Rain Removal

Adapting Object Detectors via Selective Cross-Domain Alignment

Learning a Unified Classifier Incrementally via Rebalancing

Multi-Task Self-Supervised Object Detection via Recycling of Bounding Box Annotations

Unsupervised Domain Adaptation to Improve Image Segmentation Quality Both in the Source and Target Domain

Large Scale Incremental Learning

Diversify and Match: A Domain Adaptive Representation Learning Paradigm for Object Detection

Domain Agnostic Learning with Disentangled Representations

A Comprehensive Overhaul of Feature Distillation

Strong-Weak Distribution Alignment for Adaptive Object Detection

Unsupervised Multi-Target Domain Adaptation: An Information Theoretic Approach

Learning to Learn without Forgetting By Maximizing Transfer and Minimizing Interference

UMAP: Uniform Manifold Approximation and Projection

Object Detection With Deep Learning: A Review

Progress & Compress: A scalable framework for continual learning

Cross-Domain Weakly-Supervised Object Detection Through Progressive Domain Adaptation

Domain Adaptive Faster R-CNN for Object Detection in the Wild

UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

WILDTRACK: A Multi-camera HD Dataset for Dense Unscripted Pedestrian Detection

Incremental Learning in Deep Convolutional Neural Networks Using Partial Network Sharing

Semantic Foggy Scene Understanding with Synthetic Data

Incremental Learning of Object Detectors without Catastrophic Forgetting

Focal Loss for Dense Object Detection

Conditional Adversarial Domain Adaptation

Continual Learning with Deep Generative Replay

Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks

Continual Learning Through Synaptic Intelligence

Adversarial Discriminative Domain Adaptation

Overcoming catastrophic forgetting in neural networks

Expert Gate: Lifelong Learning with a Network of Experts

Learning without Forgetting

The Cityscapes Dataset for Semantic Urban Scene Understanding

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Explaining and Harnessing Adversarial Examples

Unsupervised Domain Adaptation by Backpropagation

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Overcoming catastrophic forgetting in neural networks

Visualizing Data using t-SNE

International Journal of Computer Vision manuscript No. (will be inserted by the editor) The PASCAL Visual Object Classes (VOC) Challenge

Field of Study

Journal Information

Name

Page

Volume

Venue Information

Name

Type

URL

Alternate Names