V-SENSE Seminar on: Deep Learning for Visual Computing

31st July 2019
V-SENSE Seminar on: Deep Learning for Visual Computing


Event details:

  • Date: Thursday, 19 September, 2019
  • Time: 09.30-13:00
  • Venue: Regent House, Front Square, Trinity College Dublin – College maps
  • Registration: Fully booked.
  • A live stream of this seminar will be available to watch here on the V-SENSE YouTube Channel.  Thank you all for your registrations! We arranged for live streaming and the video will be made available online afterwards. To view the stream, link to our YouTube home page here.  Once the stream is live, it will appear on the page.• Date: Thursday, 19 September, 2019. Time: 09.30-11:00-ish.


Artificial Intelligence (AI) has made it from science fiction into everyday life. Machine Learning (ML) enabled breakthroughs due to availability of massive data and computational resources. Deep Learning (DL) in particular disrupted all areas of visual computing (VC), including computer vision/graphics and image/video processing.

The V-SENSE team of Trinity College Dublin adopted this challenge and opportunity, and made a number of significant contributions to the field of DL for VC over the last 2 years, which were published at different venues.

The goal of this seminar is to share this work (published or accepted papers) in combined form with the local academic and industrial communities, and to stimulate further discussion and collaboration.



  • 2 hours lecture session with short presentations (5-10 min) of all papers
  • Poster session and discussion with individual researchers


Topic include:

  •  Deep Normal Estimation
  • AlphaGAN: Natural Image Matting
  • DeepStereoBrush: Interactive Depth Map Creation
  • Egocentric Gesture Recognition
  • SalNet360: Saliency Estimation for 360 VR Images
  • Super-resolution of 360 VR Images
  • Deep Learning for Generating Ambisonics in 360 VR Video
  • STaDA: Neural Style Transfer for Data Augmentation
  • Classification of MoCap Data
  • Deep Lens Distortion Estimation
  • ColorNet: Estimating Colorfulness in Natural Images
  • Automated Image Aesthetics: Photographic Style Classification
  • DeepTMO: HDR Tone Mapping
  • DublinCity: Database of Semantic Labels for LiDAR Point Cloud in City Scale


All publications available here: