The PASCAL Visual Object Classes Challenge

Existing object recognition techniques often rely on human labeled data conducting to severe limitations to design a fully autonomous machine vision system. In this work, we present an intelligent machine vision system able to learn autonomously individual objects present in real environment. This system relies on salient object detection. In its design, we were inspired by early processing stages of human visual system. In this context we suggest a novel fast algorithm for visually salient object detection, robust to real-world illumination conditions. Then we use it to extract salient objects which can be efficiently used for training the machine learning-based object detection and recognition unit of the proposed system. We provide results of our salient object detection algorithm on MSRA Salient Object Database benchmark comparing its quality with other state-of-the-art approaches. The proposed system has been implemented on a humanoid robot, increasing its autonomy in learning and interaction with humans. We report and discuss the obtained results, validating the proposed concepts.

A machine learning based intelligent vision system for autonomous object detection and recognition

Building knowledge for robots can be tedious, especially if focused on object class recognition in home environments where hundreds of everyday-objects - some with a huge intra class variability - can be found. Object recognition and especially object class recognition is a key capability in home-robotics. Achieving deployable results from state-of-.the-art algorithms is not yet achievable when the number of classes increases and near real-time is the goal. Hence, we propose to exploit contextual knowledge by using sensor and hardware constraints from the robotics and home domains and show how to use the internet as a source for obtaining the required data for building a fast, vision based object categorization system for robotics. In this paper, we give an overview of the available constraints and advantages of using a robot to set priors for object classification and propose a system which covers automated model acquisition from the web, domain simulation, descriptor generation, 3D data processing from dense stereo and classification for a - not too far - robot scenario in an internet-connected home-environment. In this work we show that this system is capable of being used in home robotics in a fast and robust way for recognition of object classes commonly found in such environments, including but not limited to chairs and mugs. We also discuss challenges and missing pieces in the framework and useful extensions.

3D object classification for mobile robots in home-environments using web-data

Deploying a modeling framework for reusable robot behavior to enable informed strategies for domestic service robots

In this paper, we present a unifying approach for learning and recognition of objects in unstructured environments through exploration. Taking inspiration from how young infants learn objects, we establish four principles for object learning. First, early object detection is based on an attention mechanism detecting salient parts in the scene. Second, motion of the object allows more accurate object localization. Next, acquiring multiple observations of the object through manipulation allows a more robust representation of the object. And last, object recognition benefits from a multi-modal representation. Using these principles, we developed a unifying method including visual attention, smooth pursuit of the object, and a multi-view and multi-modal object representation. Our results indicate the effectiveness of this approach and the improvement of the system when multiple observations are acquired from active object manipulation.

/pdf/learning-and-recognition-of-objects-inspired-by-early-3a8xlhlr68.pdf

Learning and recognition of objects inspired by early cognition

This paper describes an integrated robot system, known as Curious George, that has demonstrated state-of-the-art capabilities to recognize objects in the real world. We describe the capabilities of this system, including: the ability to access web-based training data automatically and in near real-time, the ability to model the visual appearance and 3D shape of a wide variety of object categories, navigation abilities such as exploration, mapping and path following, the ability to decompose the environment based on 3D structure, allowing for attention to be focused on regions of interest, the ability to capture high-quality images of objects in the environment, and finally, the ability to correctly label those objects with high accuracy. The competence of the combined system has been validated by entry into an international competition where Curious George has been among the top performing systems each year. We discuss the implications of such successful object recognition for society, and provide several avenues for potential improvement.

/pdf/curious-george-an-integrated-visual-search-platform-1yelio7lju.pdf

Catherine Gamroth

Papers

Curious George: An Integrated Visual Search Platform