Thomas Lucas
Thomas Lucas has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Victor Wiley
STMIK Swadharma
3 shared publications Anwar Sanusi
STIE Triguna
1 shared publication- Christophe D'ALessandro
Institut Jean Le Rond D'Alembert, CNRS Sorbonne Université, France
1 shared publication - Francesc Moreno-Noguer
Institut de Robòtica i Informàtica Industrial CSIC-UPC
1 shared publication - Ginger Delmas
Institut de Robotica i Informatica Industrial CSIC-UPC Barcelona Spain
1 shared publication - Grégory Rogez
Naver Labs Europe
1 shared publication
Latest Publications
- Open Access
PoseScript: Linking 3D Human Poses and Natural Language
Natural language plays a critical role in many computer vision applications, such as image captioning, visual question answering, and cross-modal retrieval, to provide fine-grained semantic information. Unfortunately, while human pose is key to human understanding, current 3D human pose datasets lack detailed language descriptions. To address this issue, we have introduced the PoseScript dataset. This dataset pairs more than six thousand 3D human poses from AMASS with rich human-annotated descriptions of the body parts and their spatial relationships. Additionally, to increase the size of the dataset to a scale that is compatible with data-hungry learning algorithms, we have proposed an elaborate captioning process that generates automatic synthetic descriptions in natural language from given 3D keypoints. This process extracts low-level pose information, known as "posecodes", using a set of simple but generic rules on the 3D keypoints. These posecodes are then combined into higher level textual descriptions using syntactic rules. With automatic annotations, the amount of available data significantly scales up (100k), making it possible to effectively pretrain deep models for finetuning on human captions. To showcase the potential of annotated poses, we present three multi-modal learning tasks that utilize the PoseScript dataset. Firstly, we develop a pipeline that maps 3D poses and textual descriptions into a joint embedding space, allowing for cross-modal retrieval of relevant poses from large-scale datasets. Secondly, we establish a baseline for a text-conditioned model generating 3D poses. Thirdly, we present a learned process for generating pose descriptions. These applications demonstrate the versatility and usefulness of annotated poses in various tasks and pave the way for future research in the field.
Open AccessPoseScript: Linking 3D Human Poses and Natural Language
Natural language plays a critical role in many computer vision applications, such as image captioning, visual question answering, and cross-modal retrieval, to provide fine-grained semantic information. Unfortunately, while human pose is key to human understanding, current 3D human pose datasets lack detailed language descriptions. To address this issue, we have introduced the PoseScript dataset. This dataset pairs more than six thousand 3D human poses from AMASS with rich human-annotated descriptions of the body parts and their spatial relationships. Additionally, to increase the size of the dataset to a scale that is compatible with data-hungry learning algorithms, we have proposed an elaborate captioning process that generates automatic synthetic descriptions in natural language from given 3D keypoints. This process extracts low-level pose information, known as "posecodes", using a set of simple but generic rules on the 3D keypoints. These posecodes are then combined into higher level textual descriptions using syntactic rules. With automatic annotations, the amount of available data significantly scales up (100k), making it possible to effectively pretrain deep models for finetuning on human captions. To showcase the potential of annotated poses, we present three multi-modal learning tasks that utilize the PoseScript dataset. Firstly, we develop a pipeline that maps 3D poses and textual descriptions into a joint embedding space, allowing for cross-modal retrieval of relevant poses from large-scale datasets. Secondly, we establish a baseline for a text-conditioned model generating 3D poses. Thirdly, we present a learned process for generating pose descriptions. These applications demonstrate the versatility and usefulness of annotated poses in various tasks and pave the way for future research in the field.
- Open Access
- 14 July 2022
- Published by Sanata Dharma University in IJIET (International Journal of Indonesian Education and Teaching)
- Vol. 6 (2) , 226-242
THE IMPACT OF COMPUTERIZED EXAM FEATURES, EFFECTIVE EXAM IMPLEMENTATION, TOWARD STUDENT SATISFACTION
This paper intends to examine in detail the relationship between computerized exam features and effective exam implementation toward student satisfaction. The main goal is to examine the synergistic relationships which impacting student satisfaction after they using a computerized standardization exam in the college setting. Design/methodology/approach –an initial exploratory analysis was conducted methodically to confirm the proposed model through a structural equation modeling approach. The computerized exam features (especially, clear wording instruction and exam rulebook) had high effects on student satisfaction. Secondly, Effective Exam implementation (especially, efficient effort, time and cost) also impacted student satisfaction. Thirdly, the statistical testing result confirmed a joint impact of computerized exam features and effective exam implementation toward student satisfaction. The conceptual model has been validated to understand the role of both variables toward student satisfaction. However, there is lack of empirical validation and generalization of the model into wider scope due to limited time of research and a small number of the participant which suggest future research.
- 14 July 2022
- Published by Sanata Dharma University in IJIET (International Journal of Indonesian Education and Teaching)
- Vol. 6 (2) , 226-242
Open AccessTHE IMPACT OF COMPUTERIZED EXAM FEATURES, EFFECTIVE EXAM IMPLEMENTATION, TOWARD STUDENT SATISFACTION
This paper intends to examine in detail the relationship between computerized exam features and effective exam implementation toward student satisfaction. The main goal is to examine the synergistic relationships which impacting student satisfaction after they using a computerized standardization exam in the college setting. Design/methodology/approach –an initial exploratory analysis was conducted methodically to confirm the proposed model through a structural equation modeling approach. The computerized exam features (especially, clear wording instruction and exam rulebook) had high effects on student satisfaction. Secondly, Effective Exam implementation (especially, efficient effort, time and cost) also impacted student satisfaction. Thirdly, the statistical testing result confirmed a joint impact of computerized exam features and effective exam implementation toward student satisfaction. The conceptual model has been validated to understand the role of both variables toward student satisfaction. However, there is lack of empirical validation and generalization of the model into wider scope due to limited time of research and a small number of the participant which suggest future research.
- Open Access
- 19 July 2021
- Published by PubPub
Mono-Replay : a software tool for digitized sound animation
This article describes Mono-Replay, a software environment designed for sound animation. "Sound animation" in this context means musical performance based on various modes of replay and transformation of all kinds of recorded music samples. Sound animation using Mono-Replay is a two-step process, including an off-line analysis phase and on-line performance or synthesis phase. The analysis phase proceeds with time segmentation, and the set up of anchor points corresponding to temporal musical discourse parameters (notes, pulses, events). This allows, at the performance phase, for control of timing, playback position, playback speed, and a variety of spectral effects, with the help of gesture interfaces. Animation principles and software features of Mono-Replay are described. Two examples of sound animation based on beat tracking and transient detection algorithms are presented (a multi-track record of Superstition by Steve Wonder and Jeff Beck and Accidents/Harmoniques, an electroacoustic piece by Bernard Parmegiani). With the help of these two contrasted examples, the fundamental principles of “sound animation” are reviewed: parameters of musical discourse, audio file segmentation, gestural control and interaction for animation at the performance stage.
- 19 July 2021
- Published by PubPub
Open AccessMono-Replay : a software tool for digitized sound animation
This article describes Mono-Replay, a software environment designed for sound animation. "Sound animation" in this context means musical performance based on various modes of replay and transformation of all kinds of recorded music samples. Sound animation using Mono-Replay is a two-step process, including an off-line analysis phase and on-line performance or synthesis phase. The analysis phase proceeds with time segmentation, and the set up of anchor points corresponding to temporal musical discourse parameters (notes, pulses, events). This allows, at the performance phase, for control of timing, playback position, playback speed, and a variety of spectral effects, with the help of gesture interfaces. Animation principles and software features of Mono-Replay are described. Two examples of sound animation based on beat tracking and transient detection algorithms are presented (a multi-track record of Superstition by Steve Wonder and Jeff Beck and Accidents/Harmoniques, an electroacoustic piece by Bernard Parmegiani). With the help of these two contrasted examples, the fundamental principles of “sound animation” are reviewed: parameters of musical discourse, audio file segmentation, gestural control and interaction for animation at the performance stage.
- Open Access
- 31 October 2020
- Published by Universitas Gadjah Mada in IJCCS (Indonesian Journal of Computing and Cybernetics Systems)
- Vol. 14 (4) , 429-440
Twitter’s User Opinion About Master and Doctoral Degrees: A Model of Sentiment Comparison
This paper examines the opinion of student candidate about their plan to study further to master degree (S2) and doctoral degree (S3). There is lack of approach in finding public opinion about the interest of student candidate in continuing study to higher level such as master degree or doctoral degree. Through this paper, the Twitter’s user opinions are extracted using certain data mining technique to find out three sentiment types (negative, neutral, and positive) by taking the most dominant type of emotions (i.e., anger, anticipation, love, fear, joy, sadness, surprise, trust). The dataset is divided into two groups of Twitter’s users. Both datasets represent group A those opinion is about continuing study further to master degree versus group B whose continuing to doctoral degree. The groups are then divided into three types of sentiment statements about master degree versus doctoral degree. The first group is their sentiment about continuing study further to master degree with the result: (a) 109 negative tweets, 1683 neutral tweets and 131 positive tweets. For the second group (e.g., student’s sentiments about continuing to doctoral degree), it has results: (a) 421 negative tweets, 7666 neutral tweets and 1805 positive tweets. The data are tested to give accuracy value of 85%. The result of this sentiment analysis is useful as a reference for universities to understand the development of sentiments (opinion) from Twitter’s users and help the institutions to improve their reputation and quality
- 31 October 2020
- Published by Universitas Gadjah Mada in IJCCS (Indonesian Journal of Computing and Cybernetics Systems)
- Vol. 14 (4) , 429-440
Open AccessTwitter’s User Opinion About Master and Doctoral Degrees: A Model of Sentiment Comparison
This paper examines the opinion of student candidate about their plan to study further to master degree (S2) and doctoral degree (S3). There is lack of approach in finding public opinion about the interest of student candidate in continuing study to higher level such as master degree or doctoral degree. Through this paper, the Twitter’s user opinions are extracted using certain data mining technique to find out three sentiment types (negative, neutral, and positive) by taking the most dominant type of emotions (i.e., anger, anticipation, love, fear, joy, sadness, surprise, trust). The dataset is divided into two groups of Twitter’s users. Both datasets represent group A those opinion is about continuing study further to master degree versus group B whose continuing to doctoral degree. The groups are then divided into three types of sentiment statements about master degree versus doctoral degree. The first group is their sentiment about continuing study further to master degree with the result: (a) 109 negative tweets, 1683 neutral tweets and 131 positive tweets. For the second group (e.g., student’s sentiments about continuing to doctoral degree), it has results: (a) 421 negative tweets, 7666 neutral tweets and 1805 positive tweets. The data are tested to give accuracy value of 85%. The result of this sentiment analysis is useful as a reference for universities to understand the development of sentiments (opinion) from Twitter’s users and help the institutions to improve their reputation and quality
- Open Access
- 3 February 2020
- Published by Sanata Dharma University in IJIET (International Journal of Indonesian Education and Teaching)
- Vol. 4 (1) , 142-154
FACTORS AFFECTING PERFORMANCE OF COURSE CENTER MANAGEMENT IN URBAN AREAS
Observing the role of course center to provide public access to informal learning is rare in many literature body. In addition, there is lack of information about strategies implemented by course center to improve their performance in managing the course center management. The paper analyzes the factors influencing the performance of course center in Jakarta region to fulfill the learner needs. A novelty is proposed in this paper as a model to understand how the course center can improve their performance to provide informal learning and the performance of center management. The paper uses a quantitative type with purposive sampling questionaire which distributed to the tutor, course center manager, course owners and office staffs. Our analysis result showed that the course center needs integrating their strategies to maximize their function in the community. This research is useful to provide input to the government and course managers so that the course center in Jakarta can improve their performance to fulfill the learner needs.DOI: 10.24071/ijiet.2020.040112
- 3 February 2020
- Published by Sanata Dharma University in IJIET (International Journal of Indonesian Education and Teaching)
- Vol. 4 (1) , 142-154
Open AccessFACTORS AFFECTING PERFORMANCE OF COURSE CENTER MANAGEMENT IN URBAN AREAS
Observing the role of course center to provide public access to informal learning is rare in many literature body. In addition, there is lack of information about strategies implemented by course center to improve their performance in managing the course center management. The paper analyzes the factors influencing the performance of course center in Jakarta region to fulfill the learner needs. A novelty is proposed in this paper as a model to understand how the course center can improve their performance to provide informal learning and the performance of center management. The paper uses a quantitative type with purposive sampling questionaire which distributed to the tutor, course center manager, course owners and office staffs. Our analysis result showed that the course center needs integrating their strategies to maximize their function in the community. This research is useful to provide input to the government and course managers so that the course center in Jakarta can improve their performance to fulfill the learner needs.DOI: 10.24071/ijiet.2020.040112
- Open Access
- 9 February 2018
- Published by STMIK Dharma Wacana in International Journal of Artificial Intelligence Research
- Vol. 2 (1) , 22-31
Computer Vision and Image Processing: A Paper Review
Computer vision has been studied from many persective. It expands from raw data recording into techniques and ideas combining digital image processing, pattern recognition, machine learning and computer graphics. The wide usage has attracted many scholars to integrate with many disciplines and fields. This paper provide a survey of the recent technologies and theoretical concept explaining the development of computer vision especially related to image processing using different areas of their field application. Computer vision helps scholars to analyze images and video to obtain necessary information, understand information on events or descriptions, and scenic pattern. It used method of multi-range application domain with massive data analysis. This paper provides contribution of recent development on reviews related to computer vision, image processing, and their related studies. We categorized the computer vision mainstream into four group e.g., image processing, object recognition, and machine learning. We also provide brief explanation on the up-to-date information about the techniques and their performance.
- 9 February 2018
- Published by STMIK Dharma Wacana in International Journal of Artificial Intelligence Research
- Vol. 2 (1) , 22-31
Open AccessComputer Vision and Image Processing: A Paper Review
Computer vision has been studied from many persective. It expands from raw data recording into techniques and ideas combining digital image processing, pattern recognition, machine learning and computer graphics. The wide usage has attracted many scholars to integrate with many disciplines and fields. This paper provide a survey of the recent technologies and theoretical concept explaining the development of computer vision especially related to image processing using different areas of their field application. Computer vision helps scholars to analyze images and video to obtain necessary information, understand information on events or descriptions, and scenic pattern. It used method of multi-range application domain with massive data analysis. This paper provides contribution of recent development on reviews related to computer vision, image processing, and their related studies. We categorized the computer vision mainstream into four group e.g., image processing, object recognition, and machine learning. We also provide brief explanation on the up-to-date information about the techniques and their performance.