CMIC Newsletter

VOLUME I, ISSUE 5, Date July 25th, 2011

CMIC Delivers

Technology Incubations: What Value Do They Bring?
At CMIC we aim to establish our criticality to Microsoft Research (MSR) and selected product groups (PG) through our focus on enhancing, accelerating, and optimising the transfer of MSR technologies to product with a primary focus on problems with relevance to the Middle East.

On one hand, research teams usually look ahead from three to ten years in the future (sometimes even further) depending on the research field. On the other hand, product teams develop products and services to be released to the users within months to one or two years at most. Incubation teams address this temporal gap between research and product. That’s the primary role that we play at CMIC.

Read More...

Social Networking Applications: the path between the aspirations of researchers and politicians and the worries of violating moral values and crossing privacy limits
Social networking is a term loosely used nowadays in the context of technological developments for several worldwide market scenarios. It took a special focus in our region because of its role in the Middle East spring revolutions. Researchers at CMIC were pioneers in discovering the local interest and trends of social networking services in Egypt and leading Arab countries.

Read More...

Highlights

Microsoft Launches "Microsoft Afkar"
We at CMIC have the pleasure to share with the world the public release of Microsoft Afkar, which showcases the latest technologies and innovations developed in our lab. ’Afkar’ is an Arabic word that means ‘ideas’!

The site is targeted for internet users in the Middle East and Arabic regions, introducing smart tools and solutions to enrich Arabic internet content and user experience in multiple domains like

  • Multi-language content authoring
  • Web browsing
  • Translation services
  • Arabic language experience

Read More...

CMIC Research and Publications

Transliteration Mining Using Parallel Fragment Extraction
Mining of transliterations from comparable or parallel text can enhance natural language processing applications such as machine translation and cross language information retrieval. This work presents an enhanced transliteration mining technique that uses a generative graph reinforcement model to infer mappings between source and target character sequences. An initial set of mappings are learned through automatic alignment of transliteration pairs at character sequence level.

Read More...

Active Feedback for Enhancing the Construction of Panoramic Live Mobile Video Streams
Constructing a panoramic video out of multiple incoming live mobile video streams is a challenging problem. This problem involves multiple users live streaming the same scene from different angles, using their mobile phones, with the objective of constructing a panoramic video of the scene.

Read More...

Highly Efficient Human Action Recognition using Compact 2DPCA-based descriptors in the Spatial and Transform Domains

A novel algorithm for view-invariant human action recognition is presented. This approach is based on Two-Dimensional Principal Component Analysis (2DPCA) applied directly on the Motion Energy Image (MEI) or the Motion History Image (MHI) in both the spatial domain and the transform domain. This method reduces the computational complexity by a factor of at least 66, achieving the highest recognition accuracy per camera, while maintaining minimum storage requirements, compared with the most recent reports in the field.

Read More...

Improved Optimal Seam Selection Blending for Fast Video Stitching of Videos Captured From Freely Moving Devices
We investigate the problem of stitching timely synchronized video streams captured by freely moving devices. Recently, it was shown that using frame-to-frame correlation can greatly enhance the efficiency and effectiveness of video stitching algorithms. In this work, we address some of the shortcomings from previous approaches, namely the simple blending approach, causing almost a third of stitching errors and the fact that the stitching algorithm is only tested on a frame-by frame basis which does not realistically mimic the user perception of the output system quality as a complete video.

Read More...

Higher Order Potentials with Superpixel Neighbourhood (HSN) for Semantic Image Segmentation
Among the approaches for solving the semantic image segmentation problem that has proven successful is in formulating an energy minimization expressed on top of a conditional random field (CRF) over image pixels. Recently, high order potentials (cliques of size greater than 2) over superpixels have been incorporated in the CRF energy function yielding promising results. These potentials encourage pixels within the same superpixel to take the same label by penalizing inconsistent labeling within the superpixel. While some of the earlier attempts modeled higher order potentials without considering the conditional dependencies between superpixels, others modeled these dependencies at the cost of oversimplified models at higher levels

Read More...

Seamless Annotation and Enrichment Of Mobile Captured Video Streams in Real-Time
Mobile phones are becoming more and more ubiquitous with a large number of these devices having image/video capturing capabilities, connection capabilities and built-in rich sensory. This has encouraged the common user to capture more image/video content than ever before.

Read More...

Object Matching Using Feature Aggregation Over a Frame Sequence
Object instance matching is a cornerstone component in many computer vision applications such as image search, augmented reality and unsupervised tagging. The common flow in these applications is to take an input image and match it against a database of previously enrolled images of objects of interest. This is usually difficult as one needs to capture an image corresponding to an object view already present in the database, especially in the case of 3D objects with high curvature where light reflection, viewpoint change and partial occlusion can significantly alter the appearance of the captured image.

Read More...

Personalized User-cenTric Tag Recommendation for Social Bookmarking Systems
Social bookmarking systems have been growing significantly in recent years. They allow users to bookmark URLs using their own keywords, known as tags. Tags can be used later for searching and categorizing bookmarks. With the growth of social bookmarking systems, the need to automatically recommend tags increases. Amongst the most popular approaches for tag recommendation is collaborative filtering. In this work we address two main limitations of Collaborative Filtering, the first-time seen bookmarks that have not been tagged before and the cold-start users that have no sufficient history to use for recommendation. The focus of this work is to make social personalized tag recommendation for social bookmarking systems based on finding similar users and similar bookmarks.

Read More...

Collaboration with Nile University

CMIC is always keen to establish research collaborations with local research entities and universities with the goal of advancing the state-of-the-art in computer science fields and promote world class research work within the local research community.

One such effort is undertaken with Nile University in the field of computer vision; CMIC extended its fund for the “Human activity recognition in real world videos

Furthermore, CMIC has granted Nile University a research award in the area of Information Retrieval and Data Mining with focus on translation communities.