HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

No abstract

We investigate the effectiveness of combining spatial segmentation and temporal blending to improve compression ratio and rendered quality on motion capture data of rhythmic nature. Experimental results demonstrate the feasibility of our method. We also observe that visual quality can be further improved by adopting an automatic weight propagation strategy.

No abstract

Endovascular surgery is becoming widely deployed for many critical procedures, replacing invasive medical operations with long recovery times. However, there are still many challenges in improving the efficiency and safety of its usage, and reducing surgery time; namely, regular exposure to radiation, manual navigation of surgical tools, lack of 3D visualization, and lack of intelligent planning and automatic tracking of a surgical end-effector. Thus, our goal is to develop hardware and software components of a tele-operation system to alleviate the abovementioned problems. There are three specific objectives in this project: (i) to reduce the need for a surgeon to be physically next to a patient during endovascular surgery; (ii) to overcome the difficulties encountered in manual navigation; and, (iii) to improve the speed and experience of performing such surgeries. To achieve (i) we will develop an electro-mechanical interface to accurately guide mechanically controlled surgical tools from a close distance, along with a 3D visualization interface; for (ii) we will replace the current surgical tools with an "intelligent wire" controlled by the electro-mechanical system; for (iii) we will segment 3D medical images to extract precise shapes of blood vessels, following which we will perform automatic path planning for a surgical end-effector.