GCoNet+: A Stronger Group Collaborative Co-Salient Object Detector (2022-05-30T00:00:00.000000Z)

TL;DR

A novel end-to-end group collaborative learning network, termed GCoNet+, which can effectively and efficiently identify co-salient objects in natural scenes and outperforms the existing 12 cutting-edge models.

Abstract

In this paper, we present a novel end-to-end group collaborative learning network, termed GCoNet+, which can effectively and efficiently (250 fps) identify co-salient objects in natural scenes. The proposed GCoNet+ achieves the new state-of-the-art performance for co-salient object detection (CoSOD) through mining consensus representations based on the following two essential criteria: 1) intra-group compactness to better formulate the consistency among co-salient objects by capturing their inherent shared attributes using our novel group affinity module (GAM); 2) inter-group separability to effectively suppress the influence of noisy objects on the output by introducing our new group collaborating module (GCM) conditioning on the inconsistent consensus. To further improve the accuracy, we design a series of simple yet effective components as follows: i) a recurrent auxiliary classification module (RACM) promoting model learning at the semantic level; ii) a confidence enhancement module (CEM) assisting the model in improving the quality of the final predictions; and iii) a group-based symmetric triplet (GST) loss guiding the model to learn more discriminative features. Extensive experiments on three challenging benchmarks, i.e., CoCA, CoSOD3k, and CoSal2015, demonstrate that our GCoNet+ outperforms the existing 12 cutting-edge models. Code has been released at https://github.com/ZhengPeng7/GCoNet_plus.

Authors

L. Gool

68 papers

Deng-Ping Fan

33 papers

H. Fu

27 papers

TL;DR

Abstract

Authors

References134 items

Re-Thinking the Relations in Co-Saliency Detection

Adaptive Intra-Group Aggregation for Co-Saliency Detection

Unsupervised Hierarchical Semantic Segmentation with Multiview Cosegmentation and Clustering Transformers

Global-and-Local Collaborative Learning for Co-Salient Object Detection

Democracy Does Matter: Comprehensive Feature Mining for Co-Salient Object Detection

A Unified Transformer Framework for Group-Based Segmentation: Co-Segmentation, Co-Saliency Detection and Video Salient Object Detection

Real-Time Scene Text Detection With Differentiable Binarization and Adaptive Scale Fusion

PoolNet+: Exploring the Potential of Pooling for Salient Object Detection

Group-Wise Learning for Weakly Supervised Semantic Segmentation

Summarize and Search: Learning Consensus-aware Dynamic Convolution for Co-Saliency Detection

Segmenting Objects From Relational Visual Data

PVT v2: Improved baselines with Pyramid Vision Transformer

DeepACG: Co-Saliency Detection via Semantic-aware Contrast Gromov-Wasserstein Distance

CoSformer: Detecting Co-Salient Object with Transformers

Group Collaborative Learning for Co-Salient Object Detection

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions

Exploring Cross-Image Pixel Contrast for Semantic Segmentation

Salient Object Detection via Integrity Learning

Boundary-Aware Segmentation Network for Mobile and Web Applications

CycleSegNet: Object Co-Segmentation With Cycle Refinement and Region Correspondence

CoADNet: Collaborative Aggregation-and-Distribution Networks for Co-Salient Object Detection

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

Inter-Image Communication for Weakly Supervised Localization

Re-Thinking Co-Salient Object Detection

Gradient-Induced Co-Saliency Detection

CRNet: Cross-Reference Networks for Few-Shot Segmentation

Adaptive Graph Convolutional Network With Attention Graph Clustering for Co-Saliency Detection

ABCNet: Real-Time Scene Text Spotting With Adaptive Bezier-Curve Network

MAST: A Memory-Augmented Self-Supervised Tracker

Salient Object Detection via Multiple Instance Joint Re-Learning

Weakly Supervised Few-shot Object Segmentation using Co-Attention with Visual and Semantic Embeddings

Co-saliency detection via integration of multi-layer convolutional features and inter-image propagation

PyTorch: An Imperative Style, High-Performance Deep Learning Library

A Unified Multiple Graph Learning and Convolutional Network Model for Co-saliency Estimation

Co-saliency Detection Based on Hierarchical Consistency

Stacked Cross Refinement Network for Edge-Aware Salient Object Detection

Zero-Shot Video Object Segmentation via Attentive Graph Neural Networks

Group-Wise Deep Object Co-Segmentation With Co-Attention Recurrent Neural Network

Joint Learning of Saliency Detection and Weakly Supervised Semantic Segmentation

An easy-to-hard learning strategy for within-image co-saliency detection

EGNet: Edge Guidance Network for Salient Object Detection

Few-Shot Object Detection With Attention-RPN and Multi-Relation Detector

Detecting Robust Co-Saliency with Recurrent Co-Attention Neural Network

Robust Deep Co-Saliency Detection with Group Semantic

Multiple Graph Convolutional Networks for Co-Saliency Detection

BASNet: Boundary-Aware Salient Object Detection

DeepCO3: Deep Instance Co-Segmentation by Co-Peak Search and Co-Saliency Detection

Co-Saliency Detection via Mask-Guided Fully Convolutional Networks With Multi-Scale Label Smoothing

See More, Know More: Unsupervised Video Object Segmentation With Co-Attention Siamese Networks

Co-Occurrent Features in Semantic Segmentation

Self-supervised Video Representation Learning for Correspondence Flow

Salient Object Detection in the Deep Learning Era: An In-Depth Survey

No-reference synthetic image quality assessment with convolutional neural network and local image saliency

Learning Correspondence From the Cycle-Consistency of Time

Pyramid Feature Attention Network for Saliency Detection

Efficient Video Object Co-Localization With Co-Saliency Activated Tracklets

Image Co-Saliency Detection and Co-Segmentation via Progressive Joint Optimization

SiamRPN++: Evolution of Siamese Visual Tracking With Very Deep Networks

TextMountain: Accurate Scene Text Detection via Instance Segmentation

Semantic Aware Attention Based Deep Object Co-segmentation

A Feature-Adaptive Semi-Supervised Framework for Co-saliency Detection

A Unified Metric Learning-Based Framework for Co-Saliency Detection

Triplet Loss in Siamese Network for Object Tracking

Unsupervised CNN-Based Co-saliency Detection with Graphical Optimization

Reverse Attention for Salient Object Detection

Co-attention CNNs for Unsupervised Object Co-segmentation

Tracking Emerges by Colorizing Videos

Progressive Attention Guided Recurrent Network for Salient Object Detection

Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Features

Enhanced-alignment Measure for Binary Foreground Map Evaluation

Co-Saliency Detection Within a Single Image

Deep Object Co-Segmentation

Quality-Guided Fusion-Based Co-Saliency Estimation for Image Co-Segmentation and Colocalization

Supervision by Fusion: Towards Unsupervised Learning of Deep Salient Object Detector

PiCANet: Learning Pixel-Wise Contextual Attention for Saliency Detection

Structure-Measure: A New Way to Evaluate Foreground Maps