no code implementations • 18 Apr 2024 • JunRui Zhang, Mozhgan PourKeshavarz, Amir Rasouli
In this paper, we propose to incorporate richer training dynamics information into a prototypical contrastive learning framework.
no code implementations • 11 Oct 2023 • Changhe Chen, Mozhgan PourKeshavarz, Amir Rasouli
Benchmarking is a common method for evaluating trajectory prediction models for autonomous driving.
no code implementations • 27 Jun 2023 • Mozhgan PourKeshavarz, Mohammad Sabokrou, Amir Rasouli
In autonomous driving, behavior prediction is fundamental for safe motion planning, hence the security and robustness of prediction models against adversarial attacks are of paramount importance.
no code implementations • 8 Feb 2023 • Mozhgan PourKeshavarz, Shahabedin Nabavi, Mohsen Ebrahimi Moghaddam, Mehrnoush Shamsfard
Thus, we propose a stacked cross-modal feature consolidation (SCFC) attention network for image captioning in which we simultaneously consolidate cross-modal features through a novel compounding function in a multi-step reasoning fashion.
no code implementations • ICCV 2023 • Mozhgan PourKeshavarz, Changhe Chen, Amir Rasouli
More specifically, 1) we define TAROT prediction as a novel self-supervised proxy task to identify the complex heterogeneous structure of the map.
no code implementations • 22 Mar 2021 • Mozhgan PourKeshavarz, Mohammad Sabokrou
In this paper, we shed light on an on-call transfer set to provide past experiences whenever a new class arises in the data stream.