Learning toward Visual Recognition in the Wild (ICIG 2019 Special Session)

(For any questions please email learningtovr2019@gmail.com)

Visual Recognition, as a key research field in artificial intelligence, has enabled rapid progress over the past years. Learning methodologies from various perspectives have achieved remarkable performance in tasks including image recognition, image generation, visual question-answering, visual navigation, video analytics, robotics, etc. In spite of these successes, visual recognition still faces many challenges. Humans can learn new concepts with very little supervision while the most powerful deep learning techniques might fail. This motivates the machine learning direction of zero-/few-shot learning as well as meta-learning. Besides, other obstacles such as generalization problem caused by learning with multiple domains are deserved to be investigated. It is fairly difficult to build feature representations that are transferrable between source domains and target domain. It gives rise to explorations on efficient data usages and training procedures during modeling. In addition, visual recognition often inevitably involves modeling highly structured data such as semantic graphs, videos, and natural languages. How can systems be designed and deployed for large-scale representation learning with various data structures is also a prosperous research area. These challenges include neural network architecture designing, tractable algorithms for learning, theoretical bound optimizing, and more.

This is a satellite special session of the 10th International Conference on Image and Graphics (ICIG). The special session aims to bring together researchers from both academia and industry interested in addressing various aspects of learning toward visual recognition.

Relevant topics to this special session include but are not limited to:

  • Learning with image, video, graph, text, and other structured modalities
  • Learning to address low-resource scenarios
  • Meta-learning
  • Zero-shot learning / Few-shot learning
  • Deep generative modeling
  • Deep reinforcement learning / Imitation learning
  • Domain adaptation
  • Transfer learning / Multi-task learning
  • Unsupervised representation learning
  • Scalable algorithms to accelerate learning
  • Important Dates

    Paper submission deadline May 10 May 20, 2019 (anywhere in the world)
    Acceptance notificationJune 20, 2019
    Camera-ready deadlineJune 30, 2019
    RegistrationJune 20, 2019
    Conference DateAugust 23-25, 2019

    Call for Papers

    Please submit your papers via the submission website:


  • All papers will go through a strict double-blind review process.
  • With 10 to 12 pages with for technical content including figures and references.
  • Please use the Springer LNCS Proceedings Templates.
  • Accepted papers will be included in the Proceedings of ICIG 2019, and published by Springer’s LNCS series (indexed by EI).
  • More information is available at here
  • Organization

    Session Chairs

    Tianzhu ZhangUniversity of Science and Technology of China
    Baoyuan WuTencent AI Lab
    Xi PengBinghamton University
    Wei-Lun (Harry) ChaoCornell University
    Xi (Sheryl) ZhangCornell University