Multiscale Combinatorial Grouping for Image Segmentation and Object Proposal Generation

Abstract

We propose a unified approach for bottom-up hierarchical image segmentation and object proposal generation for recognition, called Multiscale Combinatorial Grouping (MCG). For this purpose, we first develop a fast normalized cuts algorithm. We then propose a high-performance hierarchical segmenter that makes effective use of multiscale information. Finally, we propose a grouping strategy that combines our multiscale regions into highly-accurate object proposals by exploring efficiently their combinatorial space. We also present Single-scale Combinatorial Grouping (SCG), a faster version of MCG that produces competitive proposals in under five second per image. We conduct an extensive and comprehensive empirical validation on the BSDS500, SegVOC12, SBD, and COCO datasets, showing that MCG produces state-of-the-art contours, hierarchical regions, and object proposals.

References in corpus (1)

• What makes for effective detection proposals?

Cited by in corpus (81)

• Learning to Segment Object Candidates
• Scalable, High-Quality Object Detection
• Dynamic Feature Integration for Simultaneous Detection of Salient Object, Edge and Skeleton
• Affinity Attention Graph Neural Network for Weakly Supervised Semantic Segmentation
• Temporal Action Detection with Structured Segment Networks
• Convolutional Oriented Boundaries
• A Focused Dynamic Attention Model for Visual Question Answering
• Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks
• Learning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour Prediction
• Video Salient Object Detection Using Spatiotemporal Deep Features
• Solve the Puzzle of Instance Segmentation in Videos: A Weakly Supervised Framework with Spatio-Temporal Collaboration
• Adaptive Morphological Reconstruction for Seeded Image Segmentation
• A novel Region of Interest Extraction Layer for Instance Segmentation
• Hierarchical Object Detection with Deep Reinforcement Learning
• Learning Video Object Segmentation from Static Images
• Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks
• Learning to Refine Object Segments
• Ask Me Anything: Free-form Visual Question Answering Based on Knowledge from External Sources
• Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation
• Bounding Box Tightness Prior for Weakly Supervised Image Segmentation
• Regularized Densely-connected Pyramid Network for Salient Instance Segmentation
• Weakly Supervised Instance Segmentation using Class Peak Response
• ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network
• What value do explicit high level concepts have in vision to language problems?
• Deep Region Hashing for Efficient Large-scale Instance Search from Images
• Simple Does It: Weakly Supervised Instance and Semantic Segmentation
• Exploiting saliency for object segmentation from image level labels
• BoxInst: High-Performance Instance Segmentation with Box Annotations
• Spatial Aggregation of Holistically-Nested Networks for Automated Pancreas Segmentation
• Budget-aware Semi-Supervised Semantic and Instance Segmentation
• Object Counting and Instance Segmentation with Image-level Supervision
• A pooling based scene text proposal technique for scene text reading in the wild
• A Novel Upsampling and Context Convolution for Image Semantic Segmentation
• ShapeMask: Learning to Segment Novel Objects by Refining Shape Priors
• Bounding boxes for weakly supervised segmentation: Global constraints get close to full supervision
• Box-driven Class-wise Region Masking and Filling Rate Guided Loss for Weakly Supervised Semantic Segmentation
• Spatial Aggregation of Holistically-Nested Convolutional Neural Networks for Automated Pancreas Localization and Segmentation
• Label-PEnet: Sequential Label Propagation and Enhancement Networks for Weakly Supervised Instance Segmentation
• Deep Extreme Cut: From Extreme Points to Object Segmentation
• Rethinking Interactive Image Segmentation: Feature Space Annotation
• PolyTransform: Deep Polygon Transformer for Instance Segmentation
• Three Birds One Stone: A General Architecture for Salient Object Segmentation, Edge Detection and Skeleton Extraction
• Self-supervised blur detection from synthetically blurred scenes
• Improved Image Boundaries for Better Video Segmentation
• Self-Balanced R-CNN for Instance Segmentation
• Contrastive Proposal Extension with LSTM Network for Weakly Supervised Object Detection
• Fusion Flow-enhanced Graph Pooling Residual Networks for Unmanned Aerial Vehicles Surveillance in Day and Night Dual Visions
• WebSeg: Learning Semantic Segmentation from Web Searches
• Weakly-supervised Instance Segmentation via Class-agnostic Learning with Salient Images
• Weakly-supervised Contrastive Learning with Quantity Prompts for Moving Infrared Small Target Detection
• PIXOR: Real-time 3D Object Detection from Point Clouds
• An Underwater Image Semantic Segmentation Method Focusing on Boundaries and a Real Underwater Scene Semantic Segmentation Dataset
• Pano2CAD: Room Layout From A Single Panorama Image
• Complete Instances Mining for Weakly Supervised Instance Segmentation
• Weakly Supervised Object Boundaries
• Learning Semantic Segmentation with Diverse Supervision
• Deep Multicameral Decoding for Localizing Unoccluded Object Instances from a Single RGB Image
• Pointly-Supervised Instance Segmentation
• Convolutional Oriented Boundaries: From Image Segmentation to High-Level Tasks
• Robust object extraction from remote sensing data
• A Benchmark Dataset and Saliency-guided Stacked Autoencoders for Video-based Salient Object Detection
• Multiple receptive fields and small-object-focusing weakly-supervised segmentation network for fast object detection
• What leads to generalization of object proposals?
• Learning Class-Agnostic Pseudo Mask Generation for Box-Supervised Semantic Segmentation
• S4Net: Single Stage Salient-Instance Segmentation
• Unsupervised Object Detection with LiDAR Clues
• Point Cloud Instance Segmentation with Semi-supervised Bounding-Box Mining
• 3D Guided Weakly Supervised Semantic Segmentation
• Weakly Supervised Instance Segmentation for Videos with Temporal Mask Consistency
• Learning Crisp Edge Detector Using Logical Refinement Network
• Revisiting Graph Construction for Fast Image Segmentation
• Deep Grouping Model for Unified Perceptual Parsing
• A High-Performance Object Proposals based on Horizontal High Frequency Signal
• Mask-guided sample selection for Semi-Supervised Instance Segmentation
• Learning Pixel Representations for Generic Segmentation
• Flip Learning: Erase to Segment
• Learning Pixel-wise Labeling from the Internet without Human Interaction
• Oracle MCG: A first peek into COCO Detection Challenges
• Opening up Open-World Tracking
• Localized Interactive Instance Segmentation
• Learning to Segment Object Candidates via Recursive Neural Networks