Beyond Local Appearance: Category Recognition from Pairwise Interactions of Simple Features

We present a discriminative shape-based algorithm for object category localization and recognition. Our method learns object models in a weakly-supervised fashion, without requiring the specification of object locations nor pixel masks in the training data. We represent object models as cliques of fully-interconnected parts, exploiting only the pairwise geometric relationships between them. The use of pairwise relationships enables our algorithm to successfully overcome several problems that are common to previously-published methods. Even though our algorithm can easily incorporate local appearance information from richer features, we purposefully do not use them in order to demonstrate that simple geometric relationships can match (or exceed) the performance of state-of-the-art object recognition algorithms.

• Publication year

  2007

• Venue

  2007 IEEE Conference on Computer Vision and Pattern Recognition

• Publication date

  2007-06-17

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1109/CVPR.2007.383091
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Groups of Adjacent Contour Segments for Object Detection

  2008cited by this paper
• Weakly Supervised Learning of Part-Based Spatial Models for Visual Object Recognition

  2006cited by this paper
• A Boundary-Fragment-Model for Object Detection

  2006cited by this paper
• Multiple Object Class Detection with a Generative Model

  2006cited by this paper
• Sparse Flexible Models of Local Features

  2006cited by this paper
• Object Detection by Contour Segment Networks

  2006cited by this paper
• Generic object recognition with boosting

  2006cited by this paper
• Efficient MAP approximation for dense energy functions

  2006cited by this paper
• A spectral technique for correspondence problems using pairwise constraints

  2005cited by this paper
• Contour-based learning for object detection

  2005cited by this paper
• Object categorization by learned universal visual dictionary

  2005cited by this paper
• Shape matching and object recognition using low distortion correspondences

  2005cited by this paper
• Hierarchical part-based visual object categorization

  2005cited by this paper
• A sparse object category model for efficient learning and exhaustive recognition

  2005cited by this paper
• Spatial priors for part-based recognition using statistical models

  2005cited by this paper
• Models for learning spatial interactions in natural images

  2004cited by this paper
• Distinctive Image Features from Scale-Invariant Keypoints

  2004cited by this paper
• Semi-supervised training of models for appearance-based statistical object detection methods

  2004cited by this paper
• Combining Top-Down and Bottom-Up Segmentation

  2004cited by this paper
• Pictorial Structures for Object Recognition

  2004cited by this paper
• Segmentation of Multiple Salient Closed Contours from Real Images

  2003cited by this paper
• Shape Matching and Object Recognition Using Shape Contexts

  2002cited by this paper
• Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data

  2001cited by this paper
• Shape Context: A New Descriptor for Shape Matching and Object Recognition

  2000cited by this paper
• Where View-based Theories Break Down: The Role of Structure in Shape Perception and Object Recognition

  2000cited by this paper
• Markov Random Field Modeling in Computer Vision

  1995cited by this paper
• Object recognition by computer - the role of geometric constraints

  1991cited by this paper
• Submitted to Ieee Transactions on Pattern Analysis and Machine Intelligence Learning to Detect Natural Image Boundaries Using Local Brightness, Color, and Texture Cues

  year unknowncited by this paper

• Fast matching via ergodic markov chain for super-large graphs

  2020cites this paper
• DaHOG-based Mobile Robot Indoor Global Localization

  2020cites this paper
• Unsupervised Learning of Graph and Hypergraph Matching

  2020cites this paper
• Multiview Co-segmentation for Wide Baseline Images using Cross-view Supervision

  2020cites this paper
• Visual graph mining for graph matching

  2019cites this paper
• Improved object recognition results using SIFT and ORB feature detector

  2019cites this paper
• Automatic Visual Inspection and Condition-Based Maintenance for Catenary

  2019cites this paper
• Multiview Cross-supervision for Semantic Segmentation

  2018cites this paper
• System-on-a-Chip (SoC)-based Hardware Acceleration for Human Action Recognition with Core Components

  2018influential citation
• Shape Detection using Geometrical Features

  2018cites this paper
• Food Recognition Based on Image Retrieval with Global Feature Descriptor

  2018cites this paper
• cvpaper.challenge in 2016: Futuristic Computer Vision through 1, 600 Papers Survey

  2017cites this paper
• Hypercolumn-array based image representation and its application to shape-based object detection

  2017cites this paper
• An Improved Transfer learning Approach for Intrusion Detection

  2017cites this paper
• Visual Graph Mining

  2017cites this paper
• Towards a Visual Story Network Using Multiple Views for Object Recognition at Different Levels of Spatiotemporal Context

  2017cites this paper
• Computer Vision Problems in 3D Plant Phenotyping

  2017cites this paper
• Factorized Graph Matching

  2016cites this paper
• High-Level Facade Image Interpretation using Marked Point Processes

  2016cites this paper
• Direction-Length Code (DLC) To Represent Binary Objects

  2016cites this paper
• Object Discovery: Soft Attributed Graph Mining

  2016cites this paper
• Tensor Power Iteration for Multi-graph Matching

  2016cites this paper
• Junction-Based Correspondence Estimation of Plant Point Cloud Data Using Subgraph Matching

  2016cites this paper
• Distributed image classification based on high-order features

  2015cites this paper
• Active skeleton for bacteria modelling

  2015cites this paper
• A 3D-grasp synthesis algorithm to grasp unknown objects based on graspable boundary and convex segments

  2015cites this paper
• Contour Guided Hierarchical Model for Shape Matching

  2015cites this paper
• Object Detection Based on Two Level Fast Matching

  2015cites this paper
• The Cross-Depiction Problem: Computer Vision Algorithms for Recognising Objects in Artwork and in Photographs

  2015cites this paper
• A Gestaltist approach to contour-based object recognition: Combining bottom-up and top-down cues

  2015cites this paper
• Cross-depiction problem: Recognition and synthesis of photographs and artwork

  2015cites this paper
• Categorical object recognition method robust to scale changes using depth data from an RGB-D sensor

  2015cites this paper
• Efficient Texture-less Object Detection for Augmented Reality Guidance

  2015cites this paper
• From RGB-D Images to RGB Images

  2015cites this paper
• Behavior Discovery and Alignment of Articulated Object Classes from Unstructured Video

  2015cites this paper
• Mining And-Or Graphs for Graph Matching and Object Discovery

  2015cites this paper
• GMWASC: Graph matching with weighted affine and sparse constraints

  2015cites this paper
• Computational Mid-Level Vision: From Border Ownership to Categorical Object Recognition

  2015cites this paper
• Thoughts on a Recursive Classifier Graph: a Multiclass Network for Deep Object Recognition

  2014cites this paper
• Image Mosaic Method Based on SIFT Features of Line Segment

  2014cites this paper
• Graduated Consistency-Regularized Optimization for Multi-graph Matching

  2014cites this paper
• Attributed Graph Mining and Matching: An Attempt to Define and Extract Soft Attributed Patterns

  2014cites this paper
• Articulated motion discovery using pairs of trajectories

  2014cites this paper
• Texture-less object recognition using contour fragment-based features with bisected local regions

  2014cites this paper
• Sparse Coding for Key Node Selection over Networks

  2014cites this paper
• Finding Matches in a Haystack: A Max-Pooling Strategy for Graph Matching in the Presence of Outliers

  2014cites this paper
• Modeling language with structured penalties. (Modélisation du langage à l'aide de pénalités structurées)

  2014cites this paper
• Common Visual Pattern Discovery via Directed Graph

  2014cites this paper
• Latent-Class Hough Forests for 3D Object Detection and Pose Estimation

  2014cites this paper
• Modeling language with structured penalties

  2014cites this paper
• Features in Concert: Discriminative Feature Selection meets Unsupervised Clustering

  2014cites this paper
• Learning Similarities for Rigid and Non-rigid Object Detection

  2014cites this paper
• Object detection using shape features

  2014cites this paper
• Spatial, Temporal and Spatio-Temporal Correspondence for Computer Vision Problems

  2014cites this paper
• Fast and Scalable Approximate Spectral Matching for Higher Order Graph Matching

  2014cites this paper
• Learning Graph Matching: Oriented to Category Modeling from Cluttered Scenes

  2013cites this paper
• Statistical part-based models for object detection in large 3D scans

  2013cites this paper
• Top-Down Bayesian Modeling and Inference for Indoor Scenes

  2013cites this paper
• AIR: An Agent for Robust Image Matching and Retrieval

  2013cites this paper
• Object Detection Using Keygraphs

  2013cites this paper
• Graph-based inference with constraints for object detection and segmentation

  2013cites this paper
• Learning Graphs to Match

  2013cites this paper
• Shape-Based Instance Detection Under Arbitrary Viewpoint

  2013cites this paper
• Feature Points Detection and Image Matching

  2013cites this paper
• Coherent Occlusion Reasoning for Instance Recognition

  2013cites this paper
• Category Modeling from Just a Single Labeling: Use Depth Information to Guide the Learning of 2D Models

  2013cites this paper
• A Mid-Level Approach to Contour-based Categorical Object Recognition

  2013cites this paper
• Complex Filters and Higher-Order Spatial Information for Image Categorization

  2013cites this paper
• Fast Pedestrian Detection Based on GPU+CPU

  2013cites this paper
• A learning algorithm for model‐based object detection

  2013influential citation
• Efficient modeling of higher order and longer range geometry statistics

  2013cites this paper
• A novel shape information based spindle positioning method with the polarized light microscopy

  2013cites this paper
• Visual Correspondence Grouping via Local Consistent Neighborhood

  2013cites this paper
• A Correlated Parts Model for Object Detection in Large 3D Scans

  2013cites this paper
• Fast and scalable approximate spectral graph matching for correspondence problems

  2013cites this paper
• Addressing Ambiguity In Object Instance Detection

  2013cites this paper
• Learning Categorical Shape from Captioned Images

  2012cites this paper
• Generalized Boundaries from Multiple Image Interpretations

  2012cites this paper
• Contour-HOG: A Stub Feature based Level Set Method for Learning Object Contour

  2012cites this paper
• Evolutionary Hough Games for coherent object detection

  2012cites this paper
• Non-rigid image alignment for object recognition

  2012cites this paper
• Real-time Learning and Detection of 3D Texture-less Objects: A Scalable Approach

  2012cites this paper
• Detecção de objetos por reconhecimento de grafos-chave

  2012cites this paper
• D-Nets: Beyond patch-based image descriptors

  2012cites this paper
• Learning Edge-Specific Kernel Functions For Pairwise Graph Matching

  2012cites this paper
• Robust Pairwise Matching of Interest Points With Complex Wavelets

  2012cites this paper
• Efficient Kernels Couple Visual Words Through Categorical Opponency

  2012cites this paper
• Database of human segmented images and its application in boundary detection

  2012cites this paper
• Shape-Based Object Detection via Boundary Structure Segmentation

  2012influential citation
• Object Recognition by Discriminative Combinations of Line Segments, Ellipses, and Appearance Features

  2012cites this paper
• Progressive graph matching: Making a move of graphs via probabilistic voting

  2012influential citation
• 3 D Object Category Pose from 2 D Images using Probabilistic 3 D Contour Models ∗

  2011cites this paper
• Detecting and Localising Multiple 3D Objects: A Fast and Scalable Approach

  2011cites this paper
• Structured sparsity through convex optimization

  2011cites this paper
• Discriminative Learning of Contour Fragments for Object Detection

  2011cites this paper
• Contributions to a fast and robust object recognition in images. (Contributions à une reconnaissance d'objet rapide et robuste en images)

  2011cites this paper
• Detecting domestic objects with ensembles of view-tuned support vector machine cascades trained on Web images

  2011cites this paper
• Optimized Implementation of Pedestrian Tracking Using Multiple Cues on GPU

  2011cites this paper
• Visual Object Recognition

  2011cites this paper
• 3D Geometric Shape Modeling by ‘3D Contour Cloud’ Reconstruction from Stereo Videos

  2011cites this paper