Pascal Mettes

2020

• [Accepted paper] ECCV 2020: "Localizing the Common Action Among a Few Videos" [pdf coming soon].
• [Accepted paper] ECCV 2020: "PointMixup: Augmentation for Point Clouds" [pdf coming soon].
• [Accepted paper] CVIU: "Open cross-domain visual search" [pdf coming soon].
• [Accepted paper] CVPR 2020: "Searching for Actions on the Hyperbole" [pdf].
• [Organization] Area chair for BMVC 2020.
• [Accepted paper] ICMR 2020 oral: "Heterogeneous Non-Local Fusion for Multimodal Activity Recognition" [pdf].
• [Accepted paper] ICMR 2020 oral: "Interactivity Proposals for Surveillance Videos" [pdf].
• [Accepted paper] ICMR 2020 oral: "Trajectory Prediction Network for Future Anticipation of Ships" [pdf].
• [Invited talk] Speaker at VISUM summer school 2020.
• [Accepted paper] ACM TOMM: "Shuffled ImageNet-Banks for Video Event Detection and Search" [pdf].
• [Vacancy] PhD position on knowledge graph of a video [closed].
• [Invited talk] Public talk at Aalto University in Finland.
• [Vacancy] PhD position on action understanding [closed].

2019

• [Invited talk] Public talk at the NVPHBV fall meeting.
• [Accepted paper] NeurIPS 2019: "Hyperspherical Prototype Networks" [pdf].
• [Accepted paper] ICCV 2019: "Counting with Focus for Free" [pdf].
• [Accepted paper] ICCV 2019: "SILCO: Show a Few Images, Localize the Common Object" [pdf].
• [Accepted paper] ICCV workshops 2019: "4-Connected Shift Residual Networks" [pdf].
• [Accepted paper] ICCV workshops 2019: "Bayesian 3D ConvNets for Action Recognition from Few Examples" [pdf].
• [Invited talk] Speaker at VISUM summer school 2019.

Localizing Actions with Objects

08-08-2018
Objects and actions are not disjoint entities. Objects are the tools for entities to act upon their environment. This blog post outlines our recent work that investigates to what extent object knowledge can help to identify and localize actions in videos. Read more...

Action Localization from Points

18-06-2018
This blog post delves into the spatio-temporal localization of actions in videos based on a new annotation paradigm, namely point supervision. Is effective action localization possible when bypassing box annotations and relying on simple point annotations during training? Read more...