Summaries of machine learning papers
This repository contains short summaries of some machine learning papers.
UNSUPERVISED LEARNING
ECCV 2018
Deep Clustering for Unsupervised Learning of Visual Features
OBJECT DETECTION
POINT CLOUD
SELF-DRIVING CARS
ECCV 2018
Deep Continuous Fusion for Multi-Sensor 3D Object Detection
AUDIO
SOUND SOURCE LOCALIZATION
ACTION RECOGNITION
SOUND SOURCE SEPARATION
SELF-SUPERVISED
ECCV 2018
Audio-Visual Scene Analysis with Self-Supervised Multisensory Features
UNCERTAINTY
ECCV 2018
Towards Realistic Predictors
OBJECT DETECTION
ECCV 2018
Acquisition of Localization Confidence for Accurate Object Detection
OBJECT DETECTION
ECCV 2018
CornerNet: Detecting Objects as Paired Keypoints
NORMALIZATION
ECCV 2018
Group Normalization
ARCHITECTURES
ATTENTION
ECCV 2018
Convolutional Networks with Adaptive Inference Graphs
ARCHITECTURES
ATTENTION
Spatial Transformer Networks (thanks, alexobednikov)LOSS FUNCTIONS
RECOGNITION
Working hard to know your neighbor’s margins: Local descriptor learning loss (thanks, alexobednikov)FACE RECOGNITION
FACES
Neural Aggregation Network for Video Face Recognition (thanks, alexobednikov)GAN
SELF-DRIVING CARS
High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
SELF-DRIVING CARS
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art
SELF-DRIVING CARS
Systematic Testing of Convolutional Neural Networks for Autonomous Driving
SELF-DRIVING CARS
SEGMENTATION
Fast Scene Understanding for Autonomous Driving
SELF-DRIVING CARS
Arguing Machines: Perception-Control System Redundancy and Edge Case Discovery in Real-World Autonomous Driving
SELF-DRIVING CARS
GAN
REINFORCEMENT
Virtual to Real Reinforcement Learning for Autonomous Driving
SELF-DRIVING CARS
End to End Learning for Self-Driving Cars
REINFORCEMENT
Rainbow: Combining Improvements in Deep Reinforcement Learning
REINFORCEMENT
Learning to Navigate in Complex Environments
GAN
Unsupervised Image-to-Image Translation Networks
RNN
Dilated Recurrent Neural Networks
OBJECT DETECTION
TRACKING
Detect to Track and Track to Detect
ARCHITECTURES
Dilated Residual Networks
OBJECT DETECTION
Feature Pyramid Networks for Object Detection
OBJECT DETECTION
SSD: Single Shot MultiBox Detector
OBJECT DETECTION
EFFICIENT NETWORKS
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
OBJECT DETECTION
Mask R-CNN
FACES
Multi-view Face Detection Using Deep Convolutional Neural Networks (aka DDFD) (thanks, arnaldog12)GAN
On the Effects of Batch and Weight Normalization in Generative Adversarial Networks
GAN
BEGAN
GAN
StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks
ACTIVATION FUNCTIONS
Self-Normalizing Neural Networks
GAN
Wasserstein GAN (aka WGAN)OBJECT DETECTION
YOLO9000: Better, Faster, Stronger (aka YOLOv2)OBJECT DETECTION
You Only Look Once: Unified, Real-Time Object Detection (aka YOLO)OBJECT DETECTION
PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
OBJECT DETECTION
R-FCN: Object Detection via Region-based Fully Convolutional Networks
OBJECT DETECTION
Faster R-CNN
OBJECT DETECTION
Fast R-CNN
OBJECT DETECTION
Rich feature hierarchies for accurate object detection and semantic segmentation (aka R-CNN)PEDESTRIANS
Ten Years of Pedestrian Detection, What Have We Learned?
NEURAL STYLE
Instance Normalization: The Missing Ingredient for Fast Stylization
HUMAN POSE ESTIMATION
Stacked Hourglass Networks for Human Pose Estimation
FACES
DeepFace: Closing the Gap to Human-Level Performance in Face Verification
TRANSLATION
Character-based Neural Machine Translation
HUMAN POSE ESTIMATION
Convolutional Pose Machines
FACES
HyperFace: A Deep Multi-task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition
FACES
Face Attribute Prediction Using Off-the-Shelf CNN Features
FACES
CMS-RCNN: Contextual Multi-Scale Region-based CNN for Unconstrained Face Detection
GAN
InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
GAN
Improved Techniques for Training GANs
ARCHITECTURES
FractalNet: Ultra-Deep Neural Networks without Residuals
OPTIMIZERS
Adam: A Method for Stochastic Optimization
GAN
RNN
Generating images with recurrent adversarial networks
GAN
Adversarially Learned Inference
ARCHITECTURES
Resnet in Resnet: Generalizing Residual Architectures
AUTOENCODERS
Rank Ordered Autoencoders
ARCHITECTURES
Wide Residual Networks
ARCHITECTURES
Identity Mappings in Deep Residual Networks
REGULARIZATION
Swapout: Learning an ensemble of deep architectures
NEURAL STYLE
Semantic Style Transfer and Turning Two-Bit Doodles into Fine Artwork
SUPERRESOLUTION
Accurate Image Super-Resolution Using Very Deep Convolutional Networks
HUMAN POSE ESTIMATION
Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation
REINFORCEMENT
Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation
COLORIZATION
Let there be Color
NEURAL STYLE
Artistic Style Transfer for Videos
REINFORCEMENT
Playing Atari with Deep Reinforcement Learning
GENERATIVE
Attend, Infer, Repeat: Fast Scene Understanding with Generative Models
ARCHITECTURES
EFFICIENT NETWORKS
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
ACTIVATION FUNCTIONS
Noisy Activation Functions
OBJECT DETECTION
IMAGE TO TEXT
DenseCap: Fully Convolutional Localization Networks for Dense Captioning
REGULARIZATION
Deep Networks with Stochastic Depth
GAN
Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
GENERATIVE
RNN
ATTENTION
DRAW A Recurrent Neural Network for Image Generation
GENERATIVE
Generative Moment Matching Networks
GENERATIVE
RNN
Pixel Recurrent Neural Networks
GAN
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
NEURAL STYLE
A Neural Algorithm for Artistic Style
NORMALIZATION
REGULARIZATION
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
ARCHITECTURES
Deep Residual Learning for Image Recognition
ACTIVATION FUNCTIONS
Fast and Accurate Deep Networks Learning By Exponential Linear Units (ELUs)
GAN
Generative Adversarial Networks
ARCHITECTURES
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
NORMALIZATION
Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks