InternIntegerized Ai-Based Video Compression Models H/F InterDigital
- Royaume-uni
- Stage
- Bac +5
- Industrie high-tech • Telecom
Summary: In this internship at the London AI Video Lab, the objective is to study fixed-point arithmetic solutions for ensuring bit-exact video compression in AI-based video codecs. Current AI-based video compression models outperform conventional codecs, like HEVC, VVC and AV1. However, AI-based video compression models are trained using floating-point arithmetic. Unfortunately, floating point arithmetic is insufficient to ensure bit-exact execution. Bit-exact execution is needed to ensure encoded bitstreams are universally decodable across any device. Fixed-point arithmetic is a potential solution to this problem. The goal of the internship is to determine a fixed-point arithmetic setup capable of ensuring bit-exactness while maintaining model performance.
This work will be seen as one step forward toward the deployment of end-to-end trained AI-based video compression models.
The goal will be to study various fixed-point arithmetic setups for layers and components of AI-based video compression models. Quantization bit-width, scaling and bias will be studied on a per component basis. The setup will be integrated into the London AI Video Lab’s end-to-end trained video compression model. The performance of the proposed solution will be evaluated and compared to existing models.
The internship will take place in the London AI Video Lab. The intern will be mentored by scientists and will be part of a research project developing end-to-end trained AI-based video compression models.
Duration: 5-6 months, starting January–April 2026.
Responsibilities
- State-of-the-art and analysis of existing solutions
- Implementation of deterministic fixed-point Deep Learning layers with varying bit-depths
- Evaluation and reporting of results
Related work
- Nagel, Markus, et al. “A white paper on neural network quantization.” arXiv preprint arXiv:2106.08295 (2021).
- Jia, Zhaoyang, et al. “Towards practical real-time neural video compression.” Proceedings of the Computer Vision and Pattern Recognition Conference. 2025.
- Li, Zhikai, Gu, Qingyi I-ViT: Integer-only Quantization for Efficient Vision Transformer Inference Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). 2023.
Expected Outcomes
Apart from the expected outcome that corresponds to the bit-exact model and its evaluation, this internship will be expected to generate patents and publications.
Qualifications
- MSc in Computer Science, Machine Learning, Mathematics, Physics or a related field
- Fluency in C++ and Python, video processing, computer vision, PyTorch
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.
Application
Candidature only onhttps://www.interdigital.com/page/careers
#J-18808-Ljbffr…