Symmetry and Group in Attribute-Object Compositions (2020-04-01T00:00:00.000000Z)

TL;DR

This paper proposes a previously ignored principle of attribute-object transformation: Symmetry and builds a transformation framework inspired by group theory, i.e. SymNet, which can be utilized for the Compositional Zero-Shot Learning task and outperforms the state-of-the-art on widely-used benchmarks.

Abstract

Attributes and objects can compose diverse compositions. To model the compositional nature of these general concepts, it is a good choice to learn them through transformations, such as coupling and decoupling. However, complex transformations need to satisfy specific principles to guarantee the rationality. In this paper, we first propose a previously ignored principle of attribute-object transformation: Symmetry. For example, coupling peeled-apple with attribute peeled should result in peeled-apple, and decoupling peeled from apple should still output apple. Incorporating the symmetry principle, a transformation framework inspired by group theory is built, i.e. SymNet. SymNet consists of two modules, Coupling Network and Decoupling Network. With the group axioms and symmetry property as objectives, we adopt Deep Neural Networks to implement SymNet and train it in an end-to-end paradigm. Moreover, we propose a Relative Moving Distance (RMD) based recognition method to utilize the attribute change instead of the attribute pattern itself to classify attributes. Our symmetry learning can be utilized for the Compositional Zero-Shot Learning task and outperforms the state-of-the-art on widely-used benchmarks. Code is available at https://github.com/DirtyHarryLYL/SymNet.

Authors

Yong-Lu Li

5 papers

Yue Xu

3 papers

Cewu Lu

8 papers

TL;DR

Abstract

Authors

References43 items

Zero-Shot Video Object Segmentation via Attentive Graph Neural Networks

InstaBoost: Boosting Instance Segmentation via Probability Map Guided Copy-Pasting

Recognizing Unseen Attribute-Object Pair with Generative Model

Task-Driven Modular Networks for Zero-Shot Compositional Learning

HAKE: Human Activity Knowledge Engine

TAFE-Net: Task-Aware Feature Embeddings for Low Shot Learning

Transferable Interactiveness Knowledge for Human-Object Interaction Detection

Joint Semantic and Latent Attribute Modelling for Cross-Class Transfer Learning

SRDA: Generating Instance Segmentation Annotation via Scanning, Reasoning and Domain Adaptation

Scene Graph Generation from Objects, Phrases and Region Captions

Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly

From Red Wine to Red Tomato: Composition with Context

Improving Person Re-identification by Attribute and Identity Learning

Mask R-CNN

Learning to Detect Human-Object Interactions

Semantic Jitter: Dense Supervision for Visual Comparisons via Synthetic Images

Beyond Holistic Object Recognition: Enriching Image Understanding with Part States

COCO Attributes: Attributes for People, Animals, and Objects

An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild

Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations

Deep Residual Learning for Image Recognition

Discovering states and transformations in image collections

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

Deep Learning Face Attributes in the Wild

Pedestrian Attribute Recognition At Far Distance

GloVe: Global Vectors for Word Representation

Inferring Analogous Attributes

BabyTalk: Understanding and Generating Simple Image Descriptions

Efficient Estimation of Word Representations in Vector Space

UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild

Im2Text: Describing Images Using 1 Million Captioned Photographs

Relative attributes

A joint learning framework for attribute models and object descriptions

Describing people: A poselet-based approach to attribute classification

Sharing features between objects and their attributes

Caltech-UCSD Birds 200

SUN database: Large-scale scene recognition from abbey to zoo

Describing objects by their attributes

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

ImageNet: A large-scale hierarchical image database

Attributes as operators: factorizing unseen attribute-object compositions

Visualizing Data using t-SNE

On this new benchmark, our model achieves 3.03% improvement on attribute accuracy and 0.66% improvement on CZSL accuracy

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names