Saeed Ranjbar Alvar

Saeed Ranjbar Alvar

I am a Staff AI Researcher at Huawei Technologies Canada. I completed my Ph.D. at Simon Fraser University (SFU) under the supervision of Prof. Ivan V. Bajić, where my research focused on the intersection of deep learning and data compression. My current research focuses on large multimodal models, particularly their general capabilities and their ability to understand and reason about real-world scenarios, including autonomous driving and embodied AI agents.

Research Experience

Large Multimodal Models: Investigating general capabilities, efficiency, and real-world understanding, with a focus on applications such as autonomous driving and embodied AI agents.
Large Language Models: Research on improving model efficiency and deployment on Neural Processing Units (NPUs).
Watermarking: Developing robust watermarking techniques for image generation models and enhancing the capacity of general-purpose dataset watermarking.
Blockchain: Designing NFT-based data marketplaces with end-to-end traceability.
AI-Based Compression: Developing multi-task image compression systems that jointly support multiple downstream tasks while maintaining high compression efficiency.
Collaborative Intelligence: Exploring edge–cloud partitioning of deep neural networks to optimize inference performance under bandwidth constraints.
Compressed Vision: Designing compression standards and feature extraction methods optimized for machine perception rather than human visual quality.
Image and Video Compression: Proposing novel spatial prediction techniques to improve the coding efficiency of modern video coding standards.

Publications

CPPO: Contrastive perception for vision language policy optimization

A. Rezaei, M. Gholami, S. Ranjbar Alvar, K. Cannons, M. A. Hossain, Z. Weimin, S. Zhou, Y. Zhang, M. Akbari
arXiv preprint, 2026
Paper Code

From segments to scenes thumbnail

From segments to scenes: Temporal understanding in autonomous driving via vision-language models

K. Cannons*, S. Ranjbar Alvar*, M. A. Hossain, A. Rezaei, M. Gholami, A. Heidarikhazaei, Z. Weimin, Y. Zhang, M. Akbari
* equal contribution
arXiv preprint, 2025
Paper Code

Divprune: Diversity-based visual token pruning for large multimodal models

S. Ranjbar Alvar, G. Singh, M. Akbari, Y. Zhang
2025 Computer Vision and Pattern Recognition Conference (CVPR)
Paper Code

AMUSE: Adaptive multi-segment encoding for dataset watermarking

S. Ranjbar Alvar, M. Akbari, D. Yue, Y. Zhang
2025 IEEE International Conference on Multimedia and Expo (ICME)
Paper Code

Lawa: Using latent space for in-generation image watermarking

A. Rezaei, M. Akbari, S. Ranjbar Alvar, A. Fatemi, Y. Zhang
2024 European Conference on Computer Vision (ECCV)
Paper Code

Compressive feature selection for remote visual multi-task inference

Saeed Ranjbar Alvar, Ivan V. Bajić
2024 IEEE International Conference on Multimedia and Expo Workshops (ICMEW)
Paper

NFT-based data marketplace with digital watermarking

S. Ranjbar Alvar, M. Akbari, D. Yue, Y. Zhang
2023 ACM SIGKDD Conference on Knowledge Discovery and Data Mining
Paper Code

License plate privacy in collaborative visual analysis of traffic scenes

Saeed Ranjbar Alvar, Korcan Uyanik, Ivan V. Bajić
2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval (MIPR)
Paper

Joint image compression and denoising via latent-space scalability

S. Ranjbar Alvar, M. Ulhaq, H Choi, I. V. Bajić
Frontiers in Signal Processing 2, 2022
Paper

Practical noise simulation for RGB images

Saeed Ranjbar Alvar, Ivan V. Bajić
arXiv preprint, 2022
Paper Code

Membership privacy protection for image translation models via adversarial knowledge distillation

S. Ranjbar Alvar, L. Wang, J. Pei, Y. Zhang
arXiv preprint arXiv:2203.05212, 2022
Paper

Pareto-optimal bit allocation for collaborative intelligence

Saeed Ranjbar Alvar, Ivan V. Bajić
IEEE Transactions on Image Processing (TIP), 2021
Paper

Previous Research

Scalable privacy in multi-task image compression

Saeed Ranjbar Alvar, Ivan V. Bajić
International Conference on Visual Communications and Image Processing (VCIP), 2021
Paper

A dataset of labelled objects on raw video sequences

H. Choi, E. Hosseini, S. Ranjbar Alvar, R. A. Cohen, I. V. Bajić
Data in brief 34, 2021
Paper

Bit allocation for multi-task collaborative intelligence

Saeed Ranjbar Alvar, Ivan V. Bajić
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020
Paper

Multi-task learning with compressible features for collaborative intelligence

Saeed Ranjbar Alvar, Ivan V. Bajić
IEEE International Conference on Image Processing (ICIP), 2019
Paper Code

FDDB-360: Face detection in 360-degree fisheye images

J. Fu, S. Ranjbar Alvar, I. V. Bajic, R. Vaughan
IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), 2019
Paper

MV-YOLO: Motion vector-aided tracking by semantic object detection

Saeed Ranjbar Alvar, Ivan V. Bajić
IEEE International Workshop on Multimedia Signal Processing (MMSP), 2018
Paper

Can you tell a face from a HEVC bitstream?

S. Ranjbar Alvar, H. Choi, I. V. Bajić
IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), 2018
Paper

Can you find a face in a HEVC bitstream?

S. Ranjbar Alvar, H. Choi, I. V. Bajić
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018
Paper

DFTS: Deep feature transmission simulator

H. Unnibhavi, H. Choi, S. Ranjbar Alvar, I. V. Bajić
Simon Fraser University (Technical Report), 2018
Paper

On lossless intra coding in HEVC with 3-tap filters

Saeed Ranjbar Alvar, Fatih Kamisli
Signal Processing: Image Communication 47, 2016
Paper

Design template by Jon Barron

Last updated: January 2, 2026