Search

Results for "temporal consistency"

FancyVideo: Towards Dynamic and Consistent Video Generation via Cross-frame Textual Guidance

arXiv · Aug 15

FancyVideo, a new video generator, introduces a Cross-frame Textual Guidance Module (CTGM) to enhance text-to-video models. CTGM uses a Temporal Information Injector and Temporal Affinity Refiner to achieve frame-specific textual guidance, improving comprehension of temporal logic. Experiments on the EvalCrafter benchmark demonstrate FancyVideo's state-of-the-art performance in generating dynamic and consistent videos, also supporting image-to-video tasks.

Modeling Complex Object Changes in Satellite Image Time-Series: Approach based on CSP and Spatiotemporal Graph

arXiv · May 24

This paper introduces a novel approach for monitoring and analyzing the evolution of complex geographic objects in satellite image time-series. The method uses a spatiotemporal graph and constraint satisfaction problems (CSP) to model and analyze object changes. Experiments on real-world satellite images from Saudi Arabian cities demonstrate the effectiveness of the proposed approach.

Making sense of space and time in video

MBZUAI · Invalid Date

MBZUAI researchers presented a new approach to video analysis at ICCV in Paris, led by Syed Talal Wasim. The approach builds on still image processing techniques like focal modulation to analyze spatial and temporal information in video separately. It aims to improve temporal aggregation while avoiding the computational complexity of transformers. Why it matters: This research advances video understanding in computer vision by offering a more efficient method for temporal modeling, crucial for applications like activity recognition and video surveillance.