This paper presents the systems used by CLIPS- IMAG laboratory. We participated to shot seg- mentation and high-level extraction tasks. We fo- cus this year on High-Level Features Extraction task, based on key frames classification. We pro- pose an original and promising framework for in- corporating contextual information (from image content) into the concept detection process. The proposed method combines local and global clas- sifiers with stacking, using SVM. We handle topo- logic and semantic contexts in concept detection performance and proposed solutions to handle the large amount of dimensions involved in classified data.