Dynamic StreetView Image Enhancement via Meteorological Overlay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital imaging technologies, such as Google StreetView, struggle to provide users with accurate, photorealistic views of environments at desired times, often lacking details and failing to match the intended weather, season, or time conditions.
Innovation Solution
A device and method that selects and modifies images based on specific view positions and time conditions, using image descriptors to create a photorealistic representation by combining existing images and meteorological data, allowing for real-time adaptation of views to match user-defined settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If 3D rendering is used to create artificial views adapted to meteorological parameters, then the view can be customized to match desired weather conditions and time, but the view lacks photorealistic details and realistic features
Solution Approach 1:
The patent merges two approaches: using real photorealistic images as a base and overlaying 3D rendered meteorological effects (rain, snow, fog, shadows) onto these images. This combination preserves the photorealistic details from actual photos while adding the adaptability to match desired weather conditions through 3D rendering effects.
Solution Approach 2:
The system introduces an intermediary layer of meteorological effect rendering between the base photorealistic image and the final displayed view. This intermediary layer consists of semi-transparent 3D rendered elements (rain streaks, snowflakes, fog layers, shadow maps) that are composited over the original image to adapt it to desired weather conditions without losing the underlying photorealistic details.
2Adaptability or versatility
If multiple images are collected and stored for different positions and orientations, then users can view environments from various angles and positions, but the system complexity and data storage requirements increase
Solution Approach 1:
The system segments the view representation into two independent components: (1) a set of base photorealistic images captured at key positions and orientations, and (2) separate 3D rendered meteorological effect layers. This segmentation allows the base images to remain static and simple to store, while the adaptive weather effects are generated on-demand through 3D rendering, reducing the need to store multiple versions of the same image for different conditions.
Solution Approach 2:
The system uses dynamic 3D rendering to generate meteorological effects in real-time based on user-selected conditions rather than storing pre-rendered images for every possible weather scenario. The 3D models of meteorological elements (rain, snow, fog) are dynamically rendered and composited onto the base images according to the desired weather parameters, providing versatility without proportional increases in stored data.
Data Source
AI summary
A user computing device, method and computer program to display a view to the user are provided. The invention allows a user to display a number of different views with photorealistic rendering of how the view should look like at a view time, based solely on view position, and descriptors of the view. In an aspect, the invention deduces from view position and view time the view descriptors (weather parameters, traffic parameters . . . ); selects, in a collection of images, an image of the view based on view position; modifies the image to match descriptors of the view (for example, by adding a rain effect if the image in the collection is under sunshine and the view should be displayed under rain). Thus, the invention allows displaying photorealistic images of how a view should look like at any time.