9 papers
QueenVIS: Rethinking Image-Only Training for Video Instance Segmentation via Query Enrichment
Arian Kheirandish, Fardin Ayar, Ehsan Javanmardi +2
Video instance segmentation (VIS) requires models to detect, segment, and track object identities across frames, and most methods enforce temporal consistency through video-level s…
How Do Diffusion Classifiers Decide? A Bias-Centric Evaluation
Saba Fathi, Fardin Ayar, Maryam Abdolali +3
Diffusion models have recently been repurposed for zero-shot classification, giving rise to diffusion classifiers that identify the best-matching text prompt by minimizing the nois…
Trust, Don't Trust, or Flip: Robust Preference-Based Reinforcement Learning with Multi-Expert Feedback
Seyed Amir Hosseini, Maryam Abdolali, Amirhosein Tavakkoli +4
Preference-based reinforcement learning (PBRL) offers a promising alternative to explicit reward engineering by learning from pairwise trajectory comparisons. However, real-world p…
Evict3R: Training-Free Token Eviction for Memory-Bounded Streaming Visual Geometry Transformers
Soroush Mahdi, Fardin Ayar, Ehsan Javanmardi +2
Streaming visual transformers like StreamVGGT achieve strong 3D perception but suffer from unbounded growth of key value (KV) memory, which limits scalability. We propose a trainin…
Training-Free Diffusion Framework for Stylized Image Generation with Identity Preservation
Mohammad Ali Rezaei, Helia Hajikazem, Saeed Khanehgir +1
Although diffusion models have demonstrated remarkable generative capabilities, existing style transfer techniques often struggle to maintain identity while achieving high-quality…
Towards Robust LiDAR Localization: Deep Learning-based Uncertainty Estimation
Minoo Dolatabadi, Fardin Ayar, Ehsan Javanmardi +2
LiDAR-based localization and SLAM often rely on iterative matching algorithms, particularly the Iterative Closest Point (ICP) algorithm, to align sensor data with pre-existing maps…