Search Results for author: Ronald Poppe

Found 18 papers, 11 papers with code

Incremental Few-Shot Instance Segmentation

1 code implementation CVPR 2021 Dan Andrei Ganea, Bas Boom, Ronald Poppe

We learn discriminative embeddings for object instances that are merged into class representatives.

Instance Segmentation Segmentation +1

Multi-Temporal Convolutions for Human Action Recognition in Videos

1 code implementation8 Nov 2020 Alexandros Stergiou, Ronald Poppe

To address this challenge, we present a novel spatio-temporal convolution block that is capable of extracting spatio-temporal patterns at multiple temporal resolutions.

Action Recognition In Videos Temporal Action Localization +1

Class Feature Pyramids for Video Explanation

1 code implementation18 Sep 2019 Alexandros Stergiou, Georgios Kapidis, Grigorios Kalliatakis, Christos Chrysoulas, Ronald Poppe, Remco Veltkamp

We demonstrate the method on six state-of-the-art 3D convolution neural networks (CNNs) on three action recognition (Kinetics-400, UCF-101, and HMDB-51) and two egocentric action recognition datasets (EPIC-Kitchens and EGTEA Gaze+).

Action Recognition Temporal Action Localization

Light Field Saliency Detection with Deep Convolutional Networks

2 code implementations19 Jun 2019 Jun Zhang, Yamei Liu, Shengping Zhang, Ronald Poppe, Meng Wang

Light field imaging presents an attractive alternative to RGB imaging because of the recording of the direction of the incoming light.

Benchmarking Saliency Detection

Analyzing Human-Human Interactions: A Survey

1 code implementation31 Jul 2018 Alexandros Stergiou, Ronald Poppe

The main challenges stem from dealing with the considerable variation in recording setting, the appearance of the people depicted and the coordinated performance of their interaction.

Action Recognition Temporal Action Localization

Compensation Sampling for Improved Convergence in Diffusion Models

1 code implementation11 Dec 2023 Hui Lu, Albert Ali Salah, Ronald Poppe

We argue that the denoising process is crucially limited by an accumulation of the reconstruction error due to an initial inaccurate reconstruction of the target data.

Denoising Facial Inpainting

TCNet: Continuous Sign Language Recognition from Trajectories and Correlated Regions

1 code implementation18 Mar 2024 Hui Lu, Albert Ali Salah, Ronald Poppe

A key challenge in continuous sign language recognition (CSLR) is to efficiently capture long-range spatial interactions over time from the video input.

Sign Language Recognition

A personal model of trumpery: Deception detection in a real-world high-stakes setting

no code implementations5 Nov 2018 Sophie van der Zee, Ronald Poppe, Alice Havrileck, Aurelien Baillon

Our results demonstrate the power of linguistic analysis in real-world deception research when applied at the individual level and provide evidence that factually incorrect tweets are not random mistakes of the sender.

Deception Detection

Twente Debate Corpus --- A Multimodal Corpus for Head Movement Analysis

no code implementations LREC 2014 Bayu Rahayudi, Ronald Poppe, Dirk Heylen

This paper introduces a multimodal discussion corpus for the study into head movement and turn-taking patterns in debates.

Egocentric Hand Track and Object-based Human Action Recognition

no code implementations2 May 2019 Georgios Kapidis, Ronald Poppe, Elsbeth van Dam, Lucas P. J. J. Noldus, Remco C. Veltkamp

We acknowledge that the presence of objects is significant for the execution of actions by humans and in general for the description of a scene.

Action Recognition Object +3

Multitask Learning to Improve Egocentric Action Recognition

no code implementations15 Sep 2019 Georgios Kapidis, Ronald Poppe, Elsbeth van Dam, Lucas Noldus, Remco Veltkamp

We employ this idea to tackle action recognition in egocentric videos by introducing additional supervised tasks.

Action Recognition

Spatio-Temporal FAST 3D Convolutions for Human Action Recognition

no code implementations30 Sep 2019 Alexandros Stergiou, Ronald Poppe

Motivated by the often distinctive temporal characteristics of actions in either horizontal or vertical direction, we introduce a novel convolution block for CNN architectures with video input.

Action Recognition Temporal Action Localization

Learning Class Regularized Features for Action Recognition

no code implementations7 Feb 2020 Alexandros Stergiou, Ronald Poppe, Remco C. Veltkamp

We show that using Class Regularization blocks in state-of-the-art CNN architectures for action recognition leads to systematic improvement gains of 1. 8%, 1. 2% and 1. 4% on the Kinetics, UCF-101 and HMDB-51 datasets, respectively.

Action Recognition

Enhancing Video Transformers for Action Understanding with VLM-aided Training

no code implementations24 Mar 2024 Hui Lu, Hu Jian, Ronald Poppe, Albert Ali Salah

The FTP framework adds four feature processors that focus on specific aspects of human action in videos: action category, action components, action description, and context information.

Action Understanding

Cannot find the paper you are looking for? You can Submit a new open access paper.