This paper introduces MOTOR, a multimodal retrieval and re-ranking approach for medical visual question answering (MedVQA) that uses grounded captions and optimal transport to capture relationships between queries and retrieved context, leveraging both textual and visual information. MOTOR identifies clinically relevant contexts to augment VLM input, achieving higher accuracy on MedVQA datasets. Empirical analysis shows MOTOR outperforms state-of-the-art methods by an average of 6.45%.
MBZUAI researchers presented a new approach to video question answering at ICCV 2023. The method leverages insights from analyzing still images to understand video content, potentially reducing the computational resources needed for training video question answering models. Guangyi Chen, Kun Zhang, and colleagues aim to apply pre-trained image models to understand video concepts. Why it matters: This research could lead to more efficient and accessible video analysis tools, benefiting fields like healthcare and security where video data is abundant.
KAUST's Visual Computing Center (VCC) is researching computer vision, image processing, and machine learning, with applications in self-driving cars, surveillance, and security. Professor Bernard Ghanem is working on teaching machines to understand visual data semantically, similar to how humans perceive the world. Self-driving cars use visual sensors to interpret traffic signals and detect obstacles, while computer vision also assists governments and corporations with security applications like facial recognition and detecting unattended luggage. Why it matters: Advancements in computer vision at KAUST can contribute to innovations in autonomous vehicles and enhance security measures in the region.
A new benchmark, LongShOTBench, is introduced for evaluating multimodal reasoning and tool use in long videos, featuring open-ended questions and diagnostic rubrics. The benchmark addresses the limitations of existing datasets by combining temporal length and multimodal richness, using human-validated samples. LongShOTAgent, an agentic system, is also presented for analyzing long videos, with both the benchmark and agent demonstrating the challenges faced by state-of-the-art MLLMs.