Zero-Shot Object Detection by Hybrid Region Embedding (2018-05-01T00:00:00.000000Z)

TL;DR

A novel approach to tackle zero-shot object detection (ZSD) where no visual training data is available for some of the target object classes is presented, where a convex combination of embeddings are used in conjunction with a detection framework.

Abstract

Object detection is considered as one of the most challenging problems in computer vision, since it requires correct prediction of both classes and locations of objects in images. In this study, we define a more difficult scenario, namely zero-shot object detection (ZSD) where no visual training data is available for some of the target object classes. We present a novel approach to tackle this ZSD problem, where a convex combination of embeddings are used in conjunction with a detection framework. For evaluation of ZSD methods, we propose a simple dataset constructed from Fashion-MNIST images and also a custom zero-shot split for the Pascal VOC detection challenge. The experimental results suggest that our method yields promising results for ZSD.

Authors

Nazli Ikizler-Cinbis

4 papers

R. G. Cinbis

7 papers

B. Demirel

1 papers

TL;DR

Abstract

Authors

References44 items

Zero-Shot Learning With Attribute Selection

Zero-Shot Object Detection

Transductive Unbiased Embedding for Zero-Shot Learning

Zero Shot Detection

Zero-Shot Object Detection: Learning to Simultaneously Recognize and Localize Novel Concepts

Zero-Shot Learning via Attribute Regression and Class Prototype Rectification

Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms

Matrix Tri-Factorization with Manifold Regularizations for Zero-Shot Learning

Low-Rank Embedded Ensemble Semantic Dictionary for Zero-Shot Learning

Attributes2Classname: A Discriminative Model for Attribute-Based Unsupervised Zero-Shot Learning

Semantic Autoencoder for Zero-Shot Learning

YOLO9000: Better, Faster, Stronger

Feature Pyramid Networks for Object Detection

G-CNN: An Iterative Grid Based Object Detector

Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks

SSD: Single Shot MultiBox Detector

DenseBox: Unifying Landmark Localization with End to End Object Detection

An embarrassingly simple approach to zero-shot learning

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

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

Predicting Deep Zero-Shot Convolutional Neural Networks Using Textual Descriptions

Fast R-CNN

How to Transfer? Zero-Shot Object Recognition via Hierarchical Transfer of Semantic Attributes

Detector discovery in the wild: Joint multiple instance and representation learning

GloVe: Global Vectors for Word Representation

Evaluation of output embeddings for fine-grained image classification

Zero-shot recognition with unreliable attributes

Large-Scale Object Classification Using Label Relation Graphs

LSDA: Large Scale Detection through Adaptation

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

The Fastest Deformable Part Model for Object Detection

OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks

Zero-Shot Learning by Convex Combination of Semantic Embeddings

DeViSE: A Deep Visual-Semantic Embedding Model

Write a Classifier: Zero-Shot Learning Using Purely Textual Descriptions

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

Label-Embedding for Attribute-Based Classification

Evaluating knowledge transfer and zero-shot learning in a large-scale setting

Large scale image annotation: learning to rank with joint word-image embeddings

ImageNet: A large-scale hierarchical image database

Learning to detect unseen object classes by between-class attribute transfer

Describing objects by their attributes

Attribute-based classification for zeroshot visual object categorization

Edinburgh Research Explorer The PASCAL Visual Object Classes (VOC) Challenge

Field of Study

Venue Information

Name

Type

URL

Alternate Names