Activez les alertes d’offres d’emploi par e-mail !
Mulipliez les invitations à des entretiens
Créez un CV sur mesure et personnalisé en fonction du poste pour multiplier vos chances.
An established industry player is seeking a motivated R&I Research Intern to explore innovative representations of 3D scenes using meshes and Gaussians. This exciting internship offers the chance to contribute to cutting-edge research in 3D geometry and compression algorithms, with opportunities for patent applications and potential publications. Join a dynamic team in a collaborative environment where your ideas can lead to significant advancements in immersive technologies. If you're passionate about computer graphics and eager to tackle real-world challenges, this internship is your gateway to making a meaningful impact in the tech world.
R&I Research Intern, 3D-mesh and 3D-Gaussians
Location: Rennes, France
Posted on: 2 Days Ago
About InterDigital
InterDigital is a global research and development company focused primarily on wireless, video, artificial intelligence (“AI”), and related technologies. We design and develop foundational technologies that enable connected, immersive experiences in a broad range of communications and entertainment products and services. We license our innovations worldwide to companies providing such products and services, including makers of wireless communications devices, consumer electronics, IoT devices, cars and other motor vehicles, and providers of cloud-based services such as video streaming. As a leader in wireless technology, our engineers have designed and developed a wide range of innovations that are used in wireless products and networks, from the earliest digital cellular systems to 5G and today’s most advanced Wi-Fi technologies. We are also a leader in video processing and video encoding/decoding technology, with a significant AI research effort that intersects with both wireless and video technologies. Founded in 1972, InterDigital is listed on Nasdaq.
Summary:
There are several ways to represent a 3D scene. For example, the use of triangle meshes is very popular in the computer graphics community. Using 3D meshes, a 3D scene can be modeled by points in 3D space, which are connected to create triangles that describe the surface of the objects in the scene. Another way is to model a scene by Gaussians dispersed in the 3D space. Such a representation has become very popular recently, as it enables a high-quality and real-time rendering of virtual views of a 3D scene acquired from real-world images.
In this internship, we plan to explore a new representation of a 3D scene, which uses one or more 3D meshes combined with 3D Gaussians. The ultimate goal would be to design a new compression algorithm for such a representation.
We envision the following outline for the work carried out during the internship:
Literature review of any existing methods for generating a 3D mesh + 3D Gaussian scene representation, and any existing compression methods for such formats.
Choice of a representation format that:
Takes advantage of the strong points of both 3D mesh and 3D Gaussian representations, in order to work better than either 3D meshes or 3D Gaussians on their own.
Is amenable to an efficient encoding of the scene geometry and attributes.
Proposal of a new compression algorithm for the chosen data format (which may either start from a modification of an existing algorithm, or be entirely new).
The intern will be working in the 3D Native Codec team, in the Immersive Standards Group of the Video Lab.
Responsibilities:
Together with the InterDigital research team, the main responsibilities of the prospective student include:
To identify existing techniques and challenges in the field related to the internship topic;
To propose novel and efficient ideas to address the challenges;
To implement and verify the proposed methods; and
To wrap up the results into one or more patent applications and/or a paper draft.
Qualifications:
Graduate student (M.Sc.) in Computer Engineering, Computer Science, Software Engineering, or related fields.
Knowledge in one or more of the following: 3D geometry, computer vision, computer graphics, image/signal processing, deep learning, C/C++, Python. Previous experience working with 3D meshes would be a plus, but is not strictly necessary.
Knowledge about compression/information theory.
Fluent English. Excellent written and verbal communication skills.
Strong team player.
Keywords: 3D, mesh, Gaussian splatting, compression, immersive
Expected Outcomes:
Contribution to/development of software modules: C++ functions, Python scripts
Patent(s)
Potential paper publication depending on results
Potential standardization contribution
Location: Rennes, France