Microsoft Research

• Abstract

  Several machine learning techniques have relied on large labeled data sets to construct predictive models and solving supervised learning tasks. The use of deep learning techniques can be highlighted, since it have been broadly and successfully used in various domains. On the other hand, in many circumstances, the labeled sets are unavailable or insufficient to train effective supervised …

  Weakly supervised learning for compressed video analysis on retrieval and classification tasks for visual alert, AP.PITE

• Abstract

  A central goal in the field of Computer Vision is image understanding. In general, appearance cues are used to detect components of interest and then spatial and hierarchical relations among these components are used to "describe" the image content at the semantic level of interest. Current deep models have reached a stage of evolution in which they are able to learn and transfer low leve…

  Understanding images and deep learning models, AP.PITE
• Characterizing and predicting biomass production in sugarcane and eucalyptus plantations in Brazil, AP.PITE

(Only some records are available in English at this moment)

• Abstract

  Multimedia recognition tasks aim to identify which kind of category a particular video belongs to, i.e., it can be either a romantic or sci-fi movie. The video may also present specific actions, such as walking in the park, sports, and a race. This project's primary goal is to compare many deep architectures for feature learning and video classification to establish whether a particular t…

  A comparative analysis of depth feature for multimedia recognition tasks, BP.IC

• Abstract

  The primary purpose of this project is to understand differences among several visual representation levels in multimedia recognition tasks and to identify possible complementary information that might be combined in a final classification stage. The main objective of this scientific initiation project is the investigation of multi-level representations in multimedia recognition tasks. Di…

  Investigation of multi-level representations in multimedia recognition tasks, BP.IC
• Support for computational environments and experiments execution: data acquisition, categorization, and maintenance, BP.TT

(Only some records are available in English at this moment)