paper-with-me

Papers

Knowledge-Guided Short-Context Action Anticipation in Human-Centric Videos

2023-09-12 · Sarthak Bhagat, Simon Stepputtis, Joseph Campbell, Katia Sycara

This work focuses on anticipating long-term human actions, particularly using short video segments, which can speed up editing workflows through improved suggestions while fostering creativity by suggesting narratives. To this end, we imbue a transformer network with a symbolic knowledge graph for action anticipation in video segments by boosting certain aspects of the transformer's attention mechanism at run-time. Demonstrated on two benchmark datasets, Breakfast and 50Salads, our approach outperforms current state-of-the-art methods for long-term action anticipation using short video context by up to 9%.

📄 PDF Abstract BibTeX arXiv:2309.05943

Code (0)

등록된 구현이 없습니다.

Tasks

Action AnticipationLong Term Action Anticipation

Methods 이 논문이 사용한 방법론

SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…

Similar Papers 제목 키워드 기반

Guided Attention for Next Active Object @ EGO4D STA Challenge

2023-05-25 · Sanket Thakur, Cigdem Beyan, Pietro Morerio, Vittorio Murino 외

In this technical report, we describe the Guided-Attention mechanism based solution for the short-term anticipation (STA) challenge for the EGO4D challenge. It combines the object detections, and the spatiotemporal featu…

ObjectShort-term Object Interaction Anticipation

Enhancing Next Active Object-based Egocentric Action Anticipation with Guided Attention

2023-05-22 · Sanket Thakur, Cigdem Beyan, Pietro Morerio, Vittorio Murino 외

Short-term action anticipation (STA) in first-person videos is a challenging task that involves understanding the next active object interactions and predicting future actions. Existing action anticipation methods have p…

Action AnticipationObjectShort-term Object Interaction Anticipation

Leveraging Next-Active Objects for Context-Aware Anticipation in Egocentric Videos

2023-08-16 · Sanket Thakur, Cigdem Beyan, Pietro Morerio, Vittorio Murino 외

Objects are crucial for understanding human-object interactions. By identifying the relevant objects, one can also predict potential future interactions or actions that may occur with these objects. In this paper, we stu…

Action AnticipationActive Object LocalizationHuman-Object Interaction DetectionObject+2

Context-Enhanced Memory-Refined Transformer for Online Action Detection

2025-03-24 · CVPR 2025 1 · Zhanzhong Pang, Fadime Sener, Angela Yao

Online Action Detection (OAD) detects actions in streaming videos using past observations. State-of-the-art OAD approaches model past observations and their interactions with an anticipated future. The past is encoded us…

Action DetectionDecoderOnline Action Detection

Semantically Guided Representation Learning For Action Anticipation

2024-07-02 · Anxhelo Diko, Danilo Avola, Bardh Prenkaj, Federico Fontana 외

Action anticipation is the task of forecasting future activity from a partially observed sequence of events. However, this task is exposed to intrinsic future uncertainty and the difficulty of reasoning upon interconnect…

Action AnticipationRepresentation Learning