Inside IAPR

Following the tradition of this newsletter, let me introduce myself briefly. I am a fourth year PhD candidate at the Swiss Federal Institute of Technology (ETH), Zurich, working on visual object recognition. Earlier, I pursued a Master’s degree at the Technical University of Munich in Germany where I carried out research in 3D object class recognition, experimenting with untextured kitchen objects like cups, pans, glasses and bottles, etc. as well as on a scene understanding system for activity recognition by detecting and tracking simultaneously hands, objects, and background landmarks in 3D.

For my PhD, I am trying to develop object class models that output more fine-grained hypotheses than just 2D bounding boxes as are currently fashionable. I have been extending 3D deformable object modeling ideas from the very early days of computer vision (in collaboration with Konrad Schindler, Michael Stark, and Bernt Schiele) so that 3D object geometry and continuous pose hypotheses can be obtained in challenging monocular images. More recently, we have been making these models robust against relatively severe occlusions and are now attempting to use them as building blocks to reconstruct and understand complicated scenes from single images. I was also a Qualcomm fellow during 2012, and will be visiting their augmented reality research center in Vienna over the summer (2013) to do some real-time computer vision on mobile chipsets.

Why BooksBooksBooks?

While journal and conference papers usually have a better ideas-to-words ratio than books, books are still a great way to convey a longer strand of ideas or introduce a new field of study to readers that otherwise would not be possible in a limited number of pages. Besides, with the huge increase in research activity in the domains which are of interest to IAPR in the recent years, books have an important role in surveying research results and bringing forward the more successful techniques for students and practitioners. I myself am an avid reader of books on machine learning, computer vision, and pattern recognition, among other topics and have always gained from good reviews (including the ones from this newsletter) and have been interested in making my own impressions public to help fellow readers. I reviewed a book for the IAPR newsletter August 2012 issue, and another one that will appear in the April issue.

Thus I am happy to be facilitating with the book reviewing process for this valuable newsletter. I invite the readers to let me know whenever they would like any particular book to be reviewed, and also if they are interested in one of the titles we have available for review (see Free Books on the Of Interest...page).