ICCV2019論文サマリ

Adversarial Representation Learning for Text-to-Image Matching
by: Shintaro Yamamoto
Text-Image Matching Vision and Language
Adversarial Fine-Grained Composition Learning for Unseen Attribute-Object Recognition
by: Kota Yoshida
Zero-Shot Learning Generative Adversarial Network Unseen Attribute-Object Pair Recognition
Object-Aware Instance Labeling for Weakly Supervised Object Detection
by: pshiko
weekly-supervised object-detection
Semi-Supervised Pedestrian Instance Synthesis and Detection With Mutual Reinforcement
by: Yamada Ryosuke
Fashion
Bridging the Domain Gap for Ground-to-Aerial Image Matching
by: Yamada Ryosuke
UAV Dataset Vehicle
Selective Sparse Sampling for Fine-Grained Image Recognition
by: Kota Yoshida
Fine-Grained Recognition attention class response map
PointCloud Saliency Maps
by: Yamada Ryosuke
Image Aesthetic Assessment Based on Pairwise Comparison A Unified Approach to Score Regression, Binary Classification, and Personalization
by: Ryota Suzuki
aesthetic assessment personalization
Cross-X Learning for Fine-Grained Visual Categorization
by: Kota Yoshida
CapsuleVOS: Semi-Supervised Video Object Segmentation Using Capsule Routing
by: Koki Madono
Capsule segmentation
GeoStyle: Discovering Fashion Trends and Events
by: Yamada Ryosuke
Fashion
Learning Rich Features at High-Speed for Single-Shot Object Detection
by: Yosuke Shinya
object detection pre-training
Bit-Flip Attack: Crushing Neural Network With Progressive Bit Search
by: Ryota Suzuki
Attack Crushing Neural Network Bit flip
Presence-Only Geographical Priors for Fine-Grained Image Classification
by: Kota Yoshida
fine-grained recognition
Evaluating Robustness of Deep Image Super-Resolution Against Adversarial Attacks
by: Sou Uchida
Super-Resolution Adversarial attack
ARGAN: Attentive Recurrent Generative Adversarial Network for Shadow Detection and Removal
by: Sou Uchida
Shadow detection Shadow removal Attention Semi-supervised learning
Dual Directed Capsule Network for Very Low Resolution Image Recognition
by: Koki Madono
CapsuleNet VLR
Fooling Network Interpretation in Image Classification
by: Kota Yoshida
Adversarial Examples Adversarial Attack
Scalable Verified Training for Provably Robust Image Classification
by: Kota Yoshida
Adversarial perturbations
Anomaly Detection in Video Sequence With Appearance-Motion Correspondence
by: Ryota Suzuki
Anomaly detection
CutMix: Regularization Strategy to Train Strong Classifiers With Localizable Features
by: Kota Yoshida
Regional dropout strategies recognition detection
Universal Adversarial Perturbation via Prior Driven Uncertainty Approximation
by: Koki Madono
Adversarial Example Universal Adversarial Perturbation
Rethinking ImageNet Pre-Training
by: Yamada Ryosuke
ImageNet Pretrain Object Detection
Hiding Video in Audio via Reversible Generative Models
by: Ryota Suzuki
Steganography Reversible Generative Models Embedding
GSLAM: A General SLAM Framework and Benchmark
by: Ryota Suzuki
SLAM platform
Gravity as a Reference for Estimating a Person's Height From Video
by: Yosuke Shinya
human height estimation human pose estimation
Escaping Plato's Cave: 3D Shape From Adversarial Rendering
by: hiroaki aizawa
3D Reconstruction GAN
Self-Supervised Representation Learning From Multi-Domain Data
by: Koki Madono
SelfSupervised adversarial
Vision-Infused Deep Audio Inpainting
by: Yukitaka Tsuchiya
Audio Inpaint GAN
Tracking Without Bells and Whistles
by: Yamada Ryosuke
Trackor Faster R-CNN
Recursive Visual Sound Separation Using Minus-Plus Net
by: Yukitaka Tsuchiya
Sound Separation
Liquid Warping GAN: A Unified Framework for Human Motion Imitation, Appearance Transfer and Novel View Synthesis
by: Yukitaka Tsuchiya
GAN Imitation Novel View Transfer
Elaborate Monocular Point and Line SLAM With Robust Initialization
by: Ryota Suzuki
indirect SLAM robust initialization
Adaptive Density Map Generation for Crowd Counting
by: Ryota Suzuki
Croud Counting
Attention-Aware Polarity Sensitive Embedding for Affective Image Retrieval
by: Ryota Suzuki
Affective Image Retrieval Embedding Attention
Zero-Shot Emotion Recognition via Affective Structural Embedding
by: Ryota Suzuki
Emotion Recognition Embedding Zero-Shot
FW-GAN: Flow-Navigated Warping GAN for Video Virtual Try-On
by: Ryota Suzuki
Virtual Try-on Fashion GAN Warping
Hierarchical Self-Attention Network for Action Localization in Videos
by: Kensho Hara
Action Localization Self-attention
NLNL: Negative Learning for Noisy Labels
by: Kensho Hara
Noisy Label
Attention-Based Autism Spectrum Disorder Screening With Privileged Modality
by: Ryota Suzuki
Saliency Autism Spectrum Disorder Screening
Memorizing Normality to Detect Anomaly: Memory-Augmented Deep Autoencoder for Unsupervised Anomaly Detection
by: Shunsuke Nakatsuka
Anomaly Detection Memory
GODS: Generalized One-Class Discriminative Subspaces for Anomaly Detection
by: Shunsuke Nakatsuka
Anomaly Detection
Drive&Act: A Multi-Modal Dataset for Fine-Grained Driver Behavior Recognition in Autonomous Vehicles
by: kensho.hara
driver activities action recognition multi-modal
Object-Aware Instance Labeling for Weakly Supervised Object Detection
by: pshiko
self-supervised semi-supervised
Generative Multi-View Human Action Recognition
by: wakamiya.t
action recognition multi-view
Clustered Object Detection in Aerial Images
by: Akihiro Nitta
object detection satellite aerial imagery
SkyScapes Fine-Grained Semantic Understanding of Aerial Scenes
by: Akihiro Nitta
segmentation semantic segmentation dataset aerial imagery
Delving Into Robust Object Detection From Unmanned Aerial Vehicles: A Deep Nuisance Disentanglement Approach
by: Akihiro Nitta
aerial imagery adversarial training feature disentanglement object detection UAV