No abstract

No abstract

A comprehensive semantic understanding of a scene is important for many applications -but in what space should diverse semantic information (e.g., objects, scene categories, material types, texture, etc.) be grounded and what should be its structure? Aspiring to have one unified structure that hosts diverse types of semantics, we follow the Scene Graph paradigm in 3D, generating a 3D Scene Graph. Given a 3D mesh and registered panoramic images, we construct a graph that spans the entire building and includes semantics on objects (e.g., class, material, and other attributes), rooms (e.g., scene category, volume, etc.) and cameras (e.g., location, etc.), as well as the relationships among these entities.However, this process is prohibitively labor heavy if done manually. To alleviate this we devise a semi-automatic framework that employs existing detection methods and enhances them using two main constraints: I. framing of query images sampled on panoramas to maximize the performance of 2D detectors, and II. multi-view consistency enforcement across 2D detections that originate in different camera locations.

What are the major findmgs of the investigation? 4 0 model development stage: It is crucial to have a complete schedule and to establish the level of detail to be used prior to 3D CAD model development for the 4D model to be used according to its origmal purpose. Carrent Izinitation~ of4D CAD: Current 4D models and 4D tools are not flexible and their applicability limited. Fatare reqzlired iqbrouements: Future improvements include facilitating 4D model generation, its manipulation and relaying detected problems to users through knowledge based systems. What outputs were generated (software, other reports, video, other)? 4D model of FDC office building, this report. 6. Research Status: What is the status of the research? Completed. What is the logcal next step? Conduct further research pertaining to the h t a t i o n s described in the report. Are the results ready to be applied or do they need further development? Both. What adktional efforts are required before this research could be applied? A more detailed 4D model, improvements of current 4D tools.