6 skills found
thanhtbt / ATT Miss[SP 2024] A Novel Recursive Least-Squares Adaptive Method For Streaming Tensor-Train Decomposition With Incomplete Observations. In Elsevier Signal Processing, 2024.
HanbaekLyu / ONMF ONTF NDLOnline Matrix/Tensor Factorization and applications to Network Dictionary Learning and images
ycq091044 / GOCPTIJCAI2022: This package provides a generalized online tensor learning tool, which maintains a low-rank CP approximation during the evolution of tensor in different ways.
ZackAkil / Online Learning With Tensorflow JsSingle page app that uses tensorflow.js to learn your actions.
20185zx / OFTDThis is the codes for "Online Functional Tensor Decomposition via Continual Learning for Streaming Data Completion"
niketworlikar / Real Time Feedback Using Facial Expression RecognitionEmotions are defined as the reflections of feelings of a Human Being. Facial Expressions play an crucial role in the recognition of emotions and are used in the process of non-verbal communication. FER can be used in various circumstances to understand the sentiments of the people. In this case, we are using FER to recognize the feedback of people attending an event or a seminar, whereas the same system can be used to detect the attentiveness of students in the classroom. Organizers of the seminar require proper feedback on their content so that they can provide even better content in the future. But traditional feedback systems such as pen and paper-based feedback or online feedback forms can’t be as trustworthy as many people just don’t bother giving feedback as it is boring and time-consuming. Also, the audience tends to give wrong and improper feedback in haste. With the help of Machine Learning and computer vision, we can develop a model that can properly identify the emotions of the people attending a seminar or an event in real-time with a high degree of accuracy in many cases. This system can also be used in the case of E-learning applications by the instructors to know how well the content was received by their audiences. In this project, we have come up with our own set of features and an algorithm for a non-traditional set of facial expressions, such as frustrated, surprise, smiling, disagree, tense and neutral. This setup not only provides the required realtime efficiency but also gives very good accuracy for these new set of gestures. The overall feedback of attendees is submitted to the instructors after the event as the system is the personindependent
