← Back to all news
arXiv Computer Vision By Sharon S. Musa, Fereshteh Forghani, Harrish Thasarathan, Sonia Joseph, Matthew Kowal, Konstantinos G. Derpanis

What, Where, and How: Probing Spatiotemporal Representations in Video Foundation Models

Read the original on arXiv Computer Vision →

The Flow has not summarised this story yet — read it at arXiv Computer Vision.

Related stories

arXiv Computer Vision
Oct 22

MS2CAST: A Video Anomaly Detection Framework via Multi-Scale Spatial Cross-Attention and Efficient Spatiotemporal Networks

MS2CAST is a semi‑supervised video anomaly detection framework that fuses multi‑scale cross‑attention between a vision transformer (MViTv2‑S) and a CNN (ConvNeXt‑T) for spatial encoding, and replaces optical flow with two interchangeable spatiotemporal modules—SwinLSTM and VMRNN—operating in latent space. The training objective combines L1 loss, Gradient Difference Loss, and Multi‑Scale SSIM to enforce pixel‑level accuracy, sharpness, and structural fidelity. Experiments on UCSD Ped2, CUHK Avenue, and ShanghaiTech show competitive ROC‑AUC scores while using less than 1 GB of GPU memory.

By Ali Safaei, Mohammad Kazemi
llmsdiffusioncomputer-vision
More like this →
arXiv Machine Learning
Sep 10

Hyperspectral Trajectory Image for Multi-Month Trajectory Anomaly Detection

The paper introduces TITAnD, a Trajectory Image Transformer that converts dense and sparse GPS trajectories into a Hyperspectral Trajectory Image (HTI) and applies vision-based classification and segmentation for anomaly detection. It employs a Cyclic Factorized Transformer (CFT) that splits attention along within-day and across-day axes, drastically reducing computational cost and enabling multi-month analysis. Empirical results show TITAnD outperforms existing sparse and dense benchmarks, achieving higher AUC-PR and faster inference than comparable Transformers.

By Md Awsafur Rahman, Chandrakanth Gudavalli, Hardik Prajapati, B. S. Manjunath
llmsagentscomputer-visionbenchmarkssafety
More like this →