Role Overview
InterDigital is seeking an intern to research novel attribute transform techniques for 3D Gaussian Splatting compression. 3D Gaussian Splatting models scenes as anisotropic Gaussian primitives with attributes such as position, scale, orientation, opacity, and appearance. The project will investigate alternatives to RAHT, including analytical, graph-based, predictive, and potentially machine learning-based transforms, to exploit correlations between Gaussian attributes and neighboring Gaussians. The goal is improved energy compaction and compression efficiency compared with RAHT while maintaining practical encoding and decoding complexity.
Responsibilities
- Analyze existing technologies and identify core challenges related to 3D Gaussian Splatting compression.
- Explore analytical, graph-based, predictive, and potentially machine learning-based transform approaches adapted to 3D Gaussian Splatting representations.
- Contribute to the design and development of software solutions addressing these challenges.
- Implement, test, and validate proposed components and algorithms.
- Perform rigorous testing and validation, including simulation-based evaluations, to ensure functional correctness and performance in line with established protocols.
- Communicate technical outcomes effectively through documentation, internal reports, publications, and intellectual property filings.
- Collaborate with the 3D Native Codec team in the New Media Codecs Group of the Video Lab.
Requirements
- Master’s student in Computer Engineering, Computer Science, Software Engineering, or a closely related discipline.
- Strong proficiency in C/C++ and Python.
- Familiarity with 3D geometry, compression, computer vision, or computer graphics is a plus.
- Fluent in English, with excellent written and verbal communication skills.
- Demonstrates strong collaboration abilities and a team-oriented mindset.