analiza aiqbench
This paper proposes DUGO, a streaming video model for detecting actions in real time without access to future frames. It combines a dual-scale encoder (local convolutions plus global self-attention) with a memory bank of stochastic Gaussian prototypes. The authors report state-of-the-art results on egocentric benchmarks (EGTEA, EPIC-Kitchens-100) and gains on Muses, but these are single-paper claims that still need independent replication, and the abstract gives no numbers. For readers, the useful takeaway is the design pattern of pairing multi-scale temporal encoding with probabilistic, input-conditioned memory for online video understanding.
DUGO: Dual-Scale Gaussian Memory for Online Temporal Action Localization in Egocentric Procedural Videos
arXiv:2610.41402v1 Announce Type: cross Abstract: Online Temporal Action Localization (OnTAL) aims to detect and localize action instances from untrimmed streaming videos in real time. Unlike offline approaches, OnTAL models have no access to future frames and cannot go back and refine earlier predictions, which makes the task considerably harder given the strict real-time constraints and narrow temporal window available. Most existing OnTAL methods lean on either brief anchor-based context or a single-scale memory of past frames, which limits how well they can capture actions unfolding across multiple time scales or requiring long-range reasoning in Procedural Videos. To address these gaps, we present DUGO (Dual-Scale Encoding and Gaussian Memory Online), a new OnTAL framework that jointly models local motion dynamics and long-range temporal dependencies. At its core, DUGO uses a Dual-Scale Temporal Encoder, pairing depth-wise local representations with global self-attention so it can pick up both fine-grained temporal patterns and broader contextual cues. We also introduce a Gaussian Memory Bank, which represents action memory as stochastic, input-conditioned Gaussian prototypes, allowing for richer temporal memory through gated cross-attention applied at both the encoder and anchor-query stages. Extensive experiments on procedural egocentric datasets (EGTEA, EPIC-Kitchens-100) demonstrate that DUGO achieves state-of-the-art performance while also providing substantial improvements on non-egocentric benchmarks such as Muses. The project page is available at: https://aman554-eq.github.io/DUGO
Abstract:Online Temporal Action Localization (OnTAL) aims to detect and localize action instances from untrimmed streaming videos in real time. Unlike offline approaches, OnTAL models have no access to future frames and cannot go back and refine earlier predictions, which makes the task considerably harder given the strict real-time constraints and narrow temporal window available. Most existing OnTAL methods lean on either brief anchor-based context or a single-scale memory of past frames, which limits how well they can capture actions unfolding across multiple time scales or requiring long-range reasoning in Procedural Videos. To address these gaps, we present DUGO (Dual-Scale Encoding and Gaussian Memory Online), a new OnTAL framework that jointly models local motion dynamics and long-range temporal dependencies. At its core, DUGO uses a Dual-Scale Temporal Encoder, pairing depth-wise local representations with global self-attention so it can pick up both fine-grained temporal patterns and broader contextual cues. We also introduce a Gaussian Memory Bank, which represents action memory as stochastic, input-conditioned Gaussian prototypes, allowing for richer temporal memory through gated cross-attention applied at both the encoder and anchor-query stages. Extensive experiments on procedural egocentric datasets (EGTEA, EPIC-Kitchens-100) demonstrate that DUGO achieves state-of-the-art performance while also providing substantial improvements on non-egocentric benchmarks such as Muses. The project page is available at: this https URL
Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
arXiv:2610.41402 [cs.CV]
(or arXiv:2610.41402v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2610.41402 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 313 cuvinte
Subiecte conexe
From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching
arXiv · 30 iul. 2026
PreAct-Nav: Agentic Reasoning Before Action for Urban Navigation
arXiv · 7 oct. 2026
EgoMoD: Predicting Global Maps of Dynamics from Local Egocentric Observations
arXiv · 12 iun. 2026
AlloSpatial: Agentic Harness Framework for Spatial Reasoning in Foundation Models
arXiv · 9 iun. 2026