Holistic, Instance-Level Human Parsing (2017-09-11T00:00:00.000000Z)

TL;DR

This work segments the parts of objects at an instance-level, such that each pixel in the image is assigned a part label, as well as the identity of the object it belongs to, and shows how this approach benefits us in obtaining segmentations at coarser granularities as well.

Abstract

Object parsing -- the task of decomposing an object into its semantic parts -- has traditionally been formulated as a category-level segmentation problem. Consequently, when there are multiple objects in an image, current methods cannot count the number of objects in the scene, nor can they determine which part belongs to which object. We address this problem by segmenting the parts of objects at an instance-level, such that each pixel in the image is assigned a part label, as well as the identity of the object it belongs to. Moreover, we show how this approach benefits us in obtaining segmentations at coarser granularities as well. Our proposed network is trained end-to-end given detections, and begins with a category-level segmentation module. Thereafter, a differentiable Conditional Random Field, defined over a variable number of instances for every input image, reasons about the identity of each part by associating it with a human detection. In contrast to other approaches, our method can handle the varying number of people in each image and our holistic network produces state-of-the-art results in instance-level part and human segmentation, together with competitive results in category-level part segmentation, all achieved by a single forward-pass through our neural network.

Authors

Philip H. S. Torr

52 papers

Anurag Arnab

3 papers

Qizhu Li

2 papers

TL;DR

Abstract

Authors

References51 items

Pixelwise Instance Segmentation with a Dynamically Instantiated Network

Look into Person: Self-Supervised Structure-Sensitive Learning and a New Benchmark for Human Parsing

Pyramid Scene Parsing Network

Learning Semantic Part-Based Models from Google Images

Bottom-up Instance Segmentation using Deep Higher-Order CRFs

DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs

Multi-scale Patch Aggregation (MPA) for Simultaneous Detection and Segmentation

End-to-End Instance Segmentation and Counting with Recurrent Attention

R-FCN: Object Detection via Region-based Fully Convolutional Networks

Training Region-Based Object Detectors with Online Hard Example Mining

Semantic Object Parsing with Graph LSTM

Instance-Level Segmentation for Autonomous Driving with Deep Densely Connected MRFs

Instance-Aware Semantic Segmentation via Multi-task Network Cascades

Deep Residual Learning for Image Recognition

Iterative Instance Segmentation

Recurrent Instance Segmentation

Higher Order Conditional Random Fields in Deep Neural Networks

Zoom Better to See Clearer: Human and Object Parsing with Hierarchical Auto-Zoom Net

Semantic Object Parsing with Local-Global Long Short-Term Memory

Reversible Recursive Instance-Level Object Segmentation

Attention to Scale: Scale-Aware Semantic Image Segmentation

Proposal-Free Network for Instance-Level Object Segmentation

Multi-instance object segmentation with occlusion handling

Monocular Object Instance Segmentation and Depth Ordering with CNNs

Joint Object and Part Segmentation Using Deep Learned Potentials

Efficient Piecewise Training of Deep Structured Models for Semantic Segmentation

Deep Human Parsing with Active Template Regression

Conditional Random Fields as Recurrent Neural Networks

Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs

Hypercolumns for object segmentation and fine-grained localization

Fully convolutional networks for semantic segmentation

Human Pose Estimation with Fields of Parts

Towards Unified Object Detection and Semantic Segmentation

Part-Based R-CNNs for Fine-Grained Category Detection

Simultaneous Detection and Segmentation

Detect What You Can: Detecting and Representing Objects Using Holistic Models and Body Parts

Human Pose Estimation Using a Joint Pixel-wise and Part-wise Formulation

Parsing clothing in fashion photographs

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials

Semantic contours from inverse detectors

Object Detection with Discriminatively Trained Part Based Models

The Layout Consistent Random Field for Recognizing and Segmenting Partially Occluded Objects

PoseCut: Simultaneous Segmentation and 3D Pose Estimation of Humans Using Dynamic Graph-Cuts

Histograms of oriented gradients for human detection

Recovering human body configurations: combining segmentation and recognition

Reﬁnenet: Multi-path reﬁnement networks with identity mappings for high-resolution semantic segmentation

Pictorial Structures for Object Recognition

Learning and example selection for object and pattern detection

The Representation and Matching of Pictorial Structures

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

Volume

Venue Information

Name

Type

URL

Alternate Names