Instance-Aware, Context-Focused, and Memory-Efficient Weakly Supervised Object Detection (2020-04-09T00:00:00.000000Z)

TL;DR

This work develops an instance-aware and context-focused unified framework for weakly supervised video object detection that achieves state-of-the-art results on COCO, VOC 2007, and VOC 2012 while devising a memory-efficient sequential batch back-propagation.

Abstract

Weakly supervised learning has emerged as a compelling tool for object detection by reducing the need for strong supervision during training. However, major challenges remain: (1) differentiation of object instances can be ambiguous; (2) detectors tend to focus on discriminative parts rather than entire objects; (3) without ground truth, object proposals have to be redundant for high recalls, causing significant memory consumption. Addressing these challenges is difficult, as it often requires to eliminate uncertainties and trivial solutions. To target these issues we develop an instance-aware and context-focused unified framework. It employs an instance-aware self-training algorithm and a learnable Concrete DropBlock while devising a memory-efficient sequential batch back-propagation. Our proposed method achieves state-of-the-art results on COCO (12.1% AP, 24.8% AP50), VOC 2007 (54.9% AP), and VOC 2012 (52.1% AP), improving baselines by great margins. In addition, the proposed method is the first to benchmark ResNet based models and weakly supervised video object detection. Refer to our project page for code, models, and more details: https://github.com/NVlabs/wetectron.

Authors

A. Schwing

14 papers

Jan Kautz

9 papers

Xiaodong Yang

12 papers

TL;DR

Abstract

Authors

References67 items

C-MIDN: Coupled Multiple Instance Detection Network With Segmentation Guidance for Weakly Supervised Object Detection

Towards Precise End-to-End Weakly Supervised Object Detection Network

WSOD2: Learning Bottom-Up and Top-Down Objectness Distillation for Weakly-Supervised Object Detection

Confidence Regularized Self-Training

You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised Object Detection

Cyclic Guidance for Weakly Supervised Joint Detection and Segmentation

C-MIL: Continuation Multiple Instance Learning for Weakly Supervised Object Detection

Weakly Supervised Object Detection With Segmentation Collaboration

Dissimilarity Coefficient Based Weakly Supervised Object Detection

DropBlock: A regularization method for convolutional networks

Weakly Supervised Region Proposal Network and Object Detection

Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-training

CornerNet: Detecting Objects as Paired Keypoints

TS2C: Tight Box Mining with Surrounding Segmentation Context for Weakly Supervised Object Detection

PCL: Proposal Cluster Learning for Weakly Supervised Object Detection

Min-Entropy Latent Model for Weakly Supervised Object Detection

W2F: A Weakly-Supervised to Fully-Supervised Framework for Object Detection

Zigzag Learning for Weakly Supervised Object Detection

Multi-evidence Filtering and Fusion for Multi-label Classification, Object Detection and Semantic Segmentation Based on Weakly Supervised Learning

Video Object Detection with an Aligned Spatial-Temporal Memory

Object Discovery By Generative Adversarial & Ranking Networks

MegDet: A Large Mini-Batch Object Detector

Memory-Efficient Implementation of DenseNets

Deep Self-Taught Learning for Weakly Supervised Object Localization

Hide-and-Seek: Forcing a Network to be Meticulous for Weakly-Supervised Object and Action Localization

A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection

Multiple Instance Detection Network with Online Instance Classifier Refinement

Flow-Guided Feature Aggregation for Video Object Detection

Object Region Mining with Adversarial Erasing: A Simple Classification to Semantic Segmentation Approach

Mask R-CNN

FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks

Weakly Supervised Cascaded Convolutional Networks

Categorical Reparameterization with Gumbel-Softmax

The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables

Temporal Ensembling for Semi-Supervised Learning

ContextLocNet: Context-Aware Deep Network Models for Weakly Supervised Localization

UberNet: Training a Universal Convolutional Neural Network for Low-, Mid-, and High-Level Vision Using Diverse Datasets and Limited Memory

Weakly Supervised Object Localization with Progressive Domain Adaptation

Training Deep Nets with Sublinear Memory Cost

Track and Transfer: Watching Videos to Simulate Strong Human Supervision for Weakly-Supervised Object Detection

Deep Residual Learning for Image Recognition

SSD: Single Shot MultiBox Detector

Weakly Supervised Deep Detection Networks

Relaxed Multiple-Instance SVM with Application to Object Discovery

You Only Look Once: Unified, Real-Time Object Detection

Variational Dropout and the Local Reparameterization Trick

Weakly supervised object detection with convex clustering

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

Fast R-CNN

Efficient object localization using Convolutional Networks

Edge Boxes: Locating Object Proposals from Edges

Weakly Supervised Object Localization with Latent Category Learning

Very Deep Convolutional Networks for Large-Scale Image Recognition

Weakly-supervised Discovery of Visual Pattern Configurations

Multiscale Combinatorial Grouping

Multi-fold MIL Training for Weakly Supervised Object Localization

Microsoft COCO: Common Objects in Context

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

Selective Search for Object Recognition

Weakly supervised object detector learning with model drift detection

ImageNet: A large-scale hierarchical image database

Learning internal representations by error propagation

Attention Networks for Weakly Supervised Object Localization

Berg . Ssd : Single shot multibox detector Megdet : A large mini - batch object detector Girshick , and Ali Farhadi . You only look once : Uni fi ed , real - time object detection

Weakly Supervised Object Detection with Posterior Regularization

Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning

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

Venue Information

Name

Type

URL

Alternate Names