Unsupervised domain adaptation for clinician pose estimation and instance segmentation in the OR (2021-08-26T00:00:00.000000Z)

TL;DR

A novel unsupervised domain adaptation method, called AdaptOR, to adapt a model from an in-the-wild labeled source domain to a statistically different unlabeled target domain, and disentangled feature normalization is proposed to handle the statistically different source and target domain data.

Authors

V. Srivastav

3 papers

A. Gangi

2 papers

Nicolas Padoy

2 papers

TL;DR

Authors

References162 items

Generative Adversarial Networks

Real-time medical phase recognition using long-term video understanding and progress gate method

FCPose: Fully Convolutional Multi-Person Pose Estimation with Dynamic Instance-Aware Convolutions

Unsupervised Domain Adaptation of Object Detectors: A Survey

Rethinking "Batch" in BatchNorm

Human pose estimation and its application to action recognition: A survey

Federated deep learning for detecting COVID-19 lung abnormalities in CT: a privacy-preserving multinational validation study

Assessment of Endovascular Team Performances Using a Comprehensive Data Capture Platform in the Hybrid Room: A Pilot Study.

OR black box and surgical control tower: Recording and streaming data and analytics to improve surgical care.

MeGA-CDA: Memory Guided Attention for Category-Aware Unsupervised Domain Adaptive Object Detection

Unbiased Teacher for Semi-Supervised Object Detection

FedBN: Federated Learning on Non-IID Features via Local Batch Normalization

Towards video-based surgical workflow understanding in open orthopaedic surgery

EvoPose2D: Pushing the Boundaries of 2D Human Pose Estimation using Neuroevolution

Surgical Data Science - from Concepts to Clinical Translation

Lownet: Privacy Preserved Ultra-Low Resolution Posture Image Classification

Collaborative Training between Region Proposal Localization and Classification for Domain Adaptive Object Detection

A Review of Single-Source Deep Unsupervised Visual Domain Adaptation

Every Pixel Matters: Center-aware Feature Alignment for Domain Adaptive Object Detector

Self-supervision on Unlabelled OR Data for Multi-person 2D/3D Human Pose Estimation

Domain Adaptation

Automatic Operating Room Surgical Activity Recognition for Robot-Assisted Surgery

Unsupervised Domain Adaptation in Semantic Segmentation: a Review

A Simple Semi-Supervised Learning Framework for Object Detection

View Invariant Human Body Detection and Pose Estimation from Multiple Depth Sensors

What Can Be Transferred: Unsupervised Domain Adaptation for Endoscopic Lesions Segmentation

A Robotic 3D Perception System for Operating Room Environment Awareness

Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation

Unbiased Mean Teacher for Cross-domain Object Detection

Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation

A Simple Framework for Contrastive Learning of Visual Representations

An Annotation Sparsification Strategy for 3D Medical Image Segmentation via Representative Selection and Self-Training

Domain Adaptive Medical Image Segmentation via Adversarial Learning of Disease-Specific Spatial Patterns

FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence

PoSeg: Pose-Aware Refinement Network for Human Instance Segmentation

PolyTransform: Deep Polygon Transformer for Instance Segmentation

Self-Supervised Learning of Pretext-Invariant Representations

Exploring uncertainty in pseudo-label guided unsupervised domain adaptation

Prior-Based Domain Adaptive Object Detection for Hazy and Rainy Conditions

Adversarial Examples Improve Image Recognition

ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring

Unsupervised Domain Adaptation via Structured Prediction Based Selective Pseudo-Labeling

DirectPose: Direct End-to-End Multi-Person Pose Estimation

Collaborative Unsupervised Domain Adaptation for Medical Image Diagnosis

CenterMask: Real-Time Anchor-Free Instance Segmentation

Momentum Contrast for Unsupervised Visual Representation Learning

A Comprehensive Survey on Transfer Learning

Chirality Nets for Human Pose Regression

CAI4CAI: The Rise of Contextual Artificial Intelligence in Computer-Assisted Interventions

Human Pose Estimation on Privacy-Preserving Low-Resolution Depth Images

SSF-DAN: Separated Semantic Feature Based Domain Adaptation Network for Semantic Segmentation

Randaugment: Practical automated data augmentation with a reduced search space

Self-Ensembling With GAN-Based Data Augmentation for Domain Adaptation in Semantic Segmentation

Self-Training and Adversarial Background Regularization for Unsupervised Domain Adaptive One-Stage Object Detection

HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation

Confidence Regularized Self-Training

Knowledge distillation for semi-supervised domain adaptation

Fusing information from multiple 2D depth cameras for 3D human pose estimation in the operating room

Automated Surgical Activity Recognition with One Labeled Sequence

Physiological synchronization and entropy as measures of team cognitive load

Data Efficient Unsupervised Domain Adaptation for Cross-Modality Image Segmentation

Cascade R-CNN: High Quality Object Detection and Instance Segmentation

Intriguing properties of adversarial training

AutoAugment: Learning Augmentation Strategies From Data

CrDoCo: Pixel-Level Domain Transfer With Cross-Domain Consistency

Domain-Specific Batch Normalization for Unsupervised Domain Adaptation

MixMatch: A Holistic Approach to Semi-Supervised Learning

Exploring Object Relation in Mean Teacher for Cross-Domain Detection

Bidirectional Learning for Domain Adaptation of Semantic Segmentation

Automatic Adaptation of Object Detectors to New Domains Using Self-Training

A Robust Learning Approach to Domain Adaptive Object Detection

FCOS: Fully Convolutional One-Stage Object Detection

TensorMask: A Foundation for Dense Object Segmentation

PifPaf: Composite Fields for Human Pose Estimation

Machine and deep learning for workflow recognition during surgery

Transformation-Consistent Self-Ensembling Model for Semisupervised Medical Image Segmentation

Deep High-Resolution Representation Learning for Human Pose Estimation