Diagram Image Retrieval and Analysis

Welcome to the DIRA workshop at CVPR 2020! (Detailed schedule on our site can be found at HERE!)

Please note that all our talks are pre-recorded (1 opening remarks, 7 keynotes, 1 contributed talk and 7 poster presentations). [We have hosted the talk videos and slides on our DIRA site to avoid heavy traffic at the CVPR landing page (i.e., this current page). For watching pre-recorded talk videos and slides, you can go HERE to find out details. However, if you are a CVPR 2020 attendee, you will need to come back to THIS CURRENT PAGE for interactions between the authors and the audience during their scheduled time slot.]

Please click the title provided in the schedule below for the corresponding talk video and slides as well as its interaction (zoom and/or live text chat) link. ENJOY!

(You are welcome to post your comments and questions under each talk. The top voted questions will be answered by the keynote/paper author(s) during  their scheduled time window. The authors will do their best to stay there for interactions during their scheduled time window through the zoom and/or live text chat link provided on their talk and slides page.)

The following THREE live Q&A sessions (June 14 in Pacific time) are mainly reserved for social networking etc.:

Teaser picture for paper
Overview of why diagram image retrieval and analysis (DIRA) matters and my recent novel DIRA work, as well as DIRA workshop summary at CVPR 2020.
    Authors: Liping Yang   
    Keywords:  Diagram images, Technical drawings, Image representation, Image classification, Line segment detection, Image retrieval, Visual similarity, Image analysis
Sund Jun14  
8:30 AM - 9:00 AM
Favorite
Teaser picture for paper
I will introduce the area of free-hand sketch analysis including fun applications as well as classic and state of the art methodologies.
    Authors: Timothy Hospedales   
    Keywords:  Free-Hand Sketch Analysis, Sketch Recognition, Sketch-Based Image Retrieval, Sketch Generation, Sketch Segmentation, Sketch Abstraction
Sund Jun14  
9:00 AM - 9:40 AM
Favorite
Teaser picture for paper
We present an interactive image search method that uses GAN-synthesized images instead of textual questions to collect relative attribute feedbacks.
    Authors: Zac Yu, Adriana Kovashka   
    Keywords:  Content-Based Image Retrieval,Interactive Image Search,Relative Attribute,Generative Adversarial Network,Image Synthesis,Image Editing,Relevance Feedback,Computer Vision,Human-Computer Interaction
Sund Jun14  
9:40 AM - 10:20 AM
Favorite
Teaser picture for paper
Visual Learning Beyond Natural Images
    Authors: Rogerio Feris   
    Keywords:  Visual Learning Beyond Natural Images, Cross-domain Few-shot Learning, Transfer Learning, Multi-Task Learning
Sund Jun14  
10:20 AM - 11:00 AM
Favorite
Teaser picture for paper
Sund Jun14  
11:00 AM - 11:20 AM
Favorite
Teaser picture for paper
A systematic review of key recent research on diagram image retrieval and analysis, with demonstration and discussion of challenges and opportunities.
    Authors: Liping Yang, Ming Gong, Vijayan K Asari   
    Keywords:  Diagram images, Technical drawings, Systematic review, Shape descriptor, CBIR, Topology and geometry, Visual similarity, Patent images
Sund Jun14  
11:20 AM - 12:20 PM
Favorite
Teaser picture for paper
We present an interactive image search method that uses GAN-synthesized images instead of textual questions to collect relative attribute feedbacks.
    Authors: Zac Yu, Adriana Kovashka   
    Keywords:  Content-Based Image Retrieval,Interactive Image Search,Relative Attribute,Generative Adversarial Network,Image Synthesis,Image Editing,Relevance Feedback,Computer Vision,Human-Computer Interaction
Sund Jun14  
11:20 AM - 12:20 PM
Favorite
Teaser picture for paper
This work presents a patent document classification/retrieval method based on image data by learning geometric shape relationships through graph CNN.
    Authors: Juan Castorena, Manish Bhattarai, Diane Oyen   
    Keywords:  Computer vision, Machine Learning, Classification, Retrieval, Graph Neural Networks, Patent Images.
Sund Jun14  
11:20 AM - 12:20 PM
Favorite
Teaser picture for paper
TSNE projection of image at various stages i) Input image space, ii) Encoder output feature space and iii) Siamese tuned output feature space.
    Authors: Manish Bhattarai, Diane Oyen, Juan Castorena, Liping Yang, Brendt Wohlberg   
    Keywords:  Diagram image retrieval, Zero-shot/One-shot learning, transfer learning, domain generalization, patent images, Scientific drawings, deep learning.
Sund Jun14  
11:20 AM - 12:20 PM
Favorite
Teaser picture for paper
A computer vision pipeline for automatically digitizing Piping and Instrumentation Diagrams (P&IDs).
    Authors: Shouvik Mani, Michael Haddad, Dan Constantini, Willy Douhard, Qiwei Li, Louis Poirier   
    Keywords:  engineering diagram, P&ID, deep learning, CNN, symbol detection, graph search, text recognition
Sund Jun14  
11:20 AM - 12:20 PM
Favorite
Teaser picture for paper
We present a method which uses scene graph embeddings as the basis for image retrieval where visual relationships are used as structured queries.
    Authors: Brigit Schroeder, Subarna Tripathi   
    Keywords:  scene graph, visual relationship, image retrieval, graph convolutional neural network, scene graph embedding
Sund Jun14  
11:20 AM - 12:20 PM
Favorite
Teaser picture for paper
We tackle the problem of zero-shot cross-modal retrieval involving color and sketch images through a novel deep representation learning technique.
    Authors: Ushasi Chaudhuri, Biplab Banerjee, Avik Bhattacharya, Datcu Mihai   
    Keywords:  Sketch, neural networks, Sketch-based image retrieval, cross-modal retrieval, Deep-learning
Sund Jun14  
11:20 AM - 12:20 PM
Favorite
Teaser picture for paper
In this talk, I described the basic idea of deep learning on graphs and introduced how to use them for question generation and grounded video descript
    Authors: Lingfei Wu   
    Keywords:  Deep Learning on Graphs, Graph Neural Networks, Natural Language Processing, Computer Vision, Question Generation, Grounded Video Description
Sund Jun14  
1:20 PM - 2:00 PM
Favorite
Teaser picture for paper
We propose to better understand creative ads with hidden messages over SOTA methods through weak multimodal supervision and common sense reasoning.
    Authors: Adriana Kovashka   
    Keywords:  visual reasoning, vision and language, common sense, visual persuasion, weak supervision
Sund Jun14  
2:00 PM - 2:45 PM
Favorite
Teaser picture for paper
Overview of why problems at the intersection of vision and language are exciting, what capabilities today's AI systems have, and what challenges remai
    Authors: Devi Parikh   
    Keywords:  Vision and language, Visual Question Answering, VQA, transformer, BERT, referring expressions, visual dialog, image captioning, demo
Sund Jun14  
2:45 PM - 3:45 PM
Favorite
Teaser picture for paper
Representing the visual world with scene graphs - a compositional, interpretable, structured knowledge representation.
    Authors: Ranjay Krishna   
    Keywords:  scene graphs, compositionality, interpretability, knowledge representations, vision and language, image retrieval, action recognition, few-shot learning, visual question answering
Sund Jun14  
3:45 PM - 4:15 PM
Favorite
Teaser picture for paper
Two case studies of interpreting and generating line drawings: occlusion-aware, example-based shape interpretation, and line drawing style translation
    Authors: William T. Freeman   
    Keywords:  line drawings, example-based methods, shape interpretation, style and content, belief propagation
Sund Jun14  
4:15 PM - 5:00 PM
Favorite
Teaser picture for paper
Sund Jun14  
5:00 PM - 5:20 PM
Favorite
Teaser picture for paper
Sund Jun14  
5:20 PM - 5:40 PM
Favorite