Researchers at MBZUAI have introduced a novel approach to enhance Large Multimodal Models (LMMs) for autonomous driving by integrating 3D tracking information. This method uses a track encoder to embed spatial and temporal data, enriching visual queries and improving the LMM's understanding of driving scenarios. Experiments on DriveLM-nuScenes and DriveLM-CARLA benchmarks demonstrate significant improvements in perception, planning, and prediction tasks compared to baseline models.
KAUST's Image and Video Understanding Lab is developing machine learning algorithms for computer vision and object tracking, with applications in video content search and UAV navigation. Their algorithms can detect specific activities in videos, helping platforms detect unwanted content and deliver relevant ads. The object tracking algorithm is also used to empower UAVs, enabling them to follow objects autonomously. Why it matters: This research enhances video content analysis and UAV capabilities, positioning KAUST as a leader in computer vision and AI applications within the region.
This article previews a talk by Gül Varol from Ecole des Ponts ParisTech on bridging natural language and 3D human motions. The talk will cover text-to-motion synthesis using generative models and text-to-motion retrieval models based on the ACTOR, TEMOS, TMR, TEACH, and SINC papers. Varol's research interests include video representation learning, human motion synthesis, and sign languages. Why it matters: Research in this area could enable more intuitive human-computer interaction and new applications in areas like virtual reality and robotics.
Video motion magnification amplifies subtle movements in video footage, making the imperceptible visible across various fields. In healthcare, it allows non-invasive monitoring of vital signs and micro-expressions. In engineering, it helps detect structural vibrations in infrastructure, while also being used in sports science, security, and robotics. Why it matters: The technology's ability to reveal hidden details has the potential to revolutionize diagnostics, monitoring, and decision-making in diverse sectors across the Middle East.
Researchers from MBZUAI have introduced VideoMolmo, a large multimodal model for spatio-temporal pointing conditioned on textual descriptions. The model incorporates a temporal module with an attention mechanism and a temporal mask fusion pipeline using SAM2 for improved coherence across video sequences. They also curated a dataset of 72k video-caption pairs and introduced VPoS-Bench, a benchmark for evaluating generalization across real-world scenarios, with code and models publicly available.
MBZUAI researchers introduce PG-Video-LLaVA, a large multimodal model with pixel-level grounding capabilities for videos, integrating audio cues for enhanced understanding. The model uses an off-the-shelf tracker and grounding module to localize objects in videos based on user prompts. PG-Video-LLaVA is evaluated on video question-answering and grounding benchmarks, using Vicuna instead of GPT-3.5 for reproducibility.
This paper introduces ADR-VINS, a monocular visual-inertial state estimation framework based on an Error-State Kalman Filter (ESKF) designed for autonomous drone racing, integrating direct pixel reprojection errors from gate corners as innovation terms. It also introduces ADR-FGO, an offline Factor-Graph Optimization framework for generating high-fidelity reference trajectories for post-flight evaluation in GNSS-denied environments. Validated on the TII-RATM dataset, ADR-VINS achieved an average RMS translation error of 0.134 m and was successfully deployed in the A2RL Drone Championship Season 2. Why it matters: The framework provides a robust and efficient solution for drone state estimation in challenging racing environments, and enables performance evaluation without relying on external localization systems.