A vision architecture for unconstrained and incremental learning of multiple categories
نویسندگان
چکیده
We present an integrated vision architecture capable of incrementally learning several visual categories based on natural hand-held objects. Additionally we focus on interactive learning, which requires real-time image processing methods and a fast learning algorithm. The overall system is composed of a figure-ground segregation part, several feature extraction methods and a life-long learning approach combining incremental learning with category-specific feature selection. In contrast to most visual categorization approaches, where typically each view is assigned to a single category, we allow labeling with an arbitrary number of shape and color categories. We also impose no restrictions on the viewing angle of presented objects, relaxing the common constraint on canonical views.
منابع مشابه
Interactive and life-long learning for identification and categorization tasks
The presented thesis focuses on life-long and interactive learning for identification and categorization tasks. The fundamental and still largely unsolved problem of life-long learning with artificial neural networks is the so-called “stability-plasticity dilemma”. To achieve plasticity the learning approach must be able to continuously integrate newly acquired knowledge into its internal repre...
متن کاملFuture-oriented implications of the resilience theory for Iran public libraries
Target: In order to play their role in social developments, public libraries face technological changes and unknown issues that can affect their identity and mission .In reference to the application of novel approaches to reconceptualize the mission of public libraries, this study tries to employ resilience theory to craft a vision for the future of Iran public libraries. Method: This study u...
متن کاملTowards Incremental Learning with Deep Convolutional Networks
Deep neural networks are a powerful class of machine learning models. However they require a lot of time and computational resources to train. We propose to apply an incremental learning approach to train models by utilizing the information present in pre-trained models. We build towards this goal by studying the relationship between network architecture, categories in training data, the amount...
متن کاملA learning and memory architecture for robot companions based on incremental associative learning
We present a learning and memory architecture that allows a robot companion to incrementally learn and associate data from different sensors and actuators. We use a topology learning algorithm that clusters the received inputs into discrete categories. On top of these clusters we apply associative learning methods to store co-occurrence relationships in an associative network. We evaluated the ...
متن کاملA Hybrid Framework for Building an Efficient Incremental Intrusion Detection System
In this paper, a boosting-based incremental hybrid intrusion detection system is introduced. This system combines incremental misuse detection and incremental anomaly detection. We use boosting ensemble of weak classifiers to implement misuse intrusion detection system. It can identify new classes types of intrusions that do not exist in the training dataset for incremental misuse detection. As...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Memetic Computing
دوره 1 شماره
صفحات -
تاریخ انتشار 2009