Towards Affordance Based Scene Understanding
نویسنده
چکیده
Scene Understanding is among the grand challenges of computer vision. We consider indoor scenes in 3D space and motivate an affordance based approach to autonomously discover common objects. Relationship between visual entities and their supported affordances is modelled as a clustering problem in a feature space reflecting human embodiment. In the current scope of work we only consider geometric properties of the visual scene. Experimental results including labelling of floor, roof, wall and estimation of scene coordinate system lay down the foundations for future work and have been presented here.
منابع مشابه
Indoor Scene Understanding in 2.5/3D: A Survey
With the availability of low-cost and compact 2.5/3D visual sensing devices, computer vision community is experiencing a growing interest in visual scene understanding. This survey paper provides a comprehensive background to this research topic. We begin with a historical perspective, followed by popular 3D data representations and a comparative analysis of available datasets. Before delving i...
متن کاملAffordance based Part Recognition for Grasping and Manipulation
Affordances (unit utility, functional and topological relationships) and semantic scene understanding are key to building a generic, scalable and cognitive architecture for visual perception. ‘Affordance based object recognition’ or recognition based on affordance features is an important step in this regard. In this paper, we extend the scope of affordance features to define ‘Conceptual Equiva...
متن کاملScene Affordance: Inferring Actions from Household Scenes
In order to be helpful some day, robots do not only need to perceive their environment accurately but also have to be capable of carrying out appropriate tasks automatically. In a household setting, it is desirable to have a robot that does not require explicit instructions but rather identifies relevant tasks automatically. We suggest to approach this challange by collecting a large dataset wh...
متن کاملWeakly Supervised Learning of Affordances
Localizing functional regions of objects or affordances is an important aspect of scene understanding. In this work, we cast the problem of affordance segmentation as that of semantic image segmentation. In order to explore various levels of supervision, we introduce a pixelannotated affordance dataset of 3090 images containing 9916 object instances with rich contextual information in terms of ...
متن کاملTowards Gestural Sonic Affordances
We present a study that explores the affordance evoked by sound and sound-gesture mappings. In order to do this, we make use of a sensor system with minimal form factor in a user study that minimizes cultural association. The present study focuses on understanding how participants describe sounds and gestures produced while playing designed sonic interaction mappings. This approach seeks to mov...
متن کامل