17945951. VIDEO GENERATION TECHNIQUES simplified abstract (NVIDIA Corporation)
Contents
- 1 VIDEO GENERATION TECHNIQUES
VIDEO GENERATION TECHNIQUES
Organization Name
Inventor(s)
Arun Mohanray Mallya of Mountain View CA (US)
Ting-Chun Wang of Santa Clara CA (US)
Ming-Yu Liu of San Jose CA (US)
VIDEO GENERATION TECHNIQUES - A simplified explanation of the abstract
This abstract first appeared for US patent application 17945951 titled 'VIDEO GENERATION TECHNIQUES
Simplified Explanation
The patent application describes apparatuses, systems, and techniques for generating a video using two or more images containing objects to be included in the video. In one embodiment, objects are identified in the images using neural networks to create a video that includes these objects.
- Objects in images are identified using neural networks.
- The identified objects are included in the generated video.
Potential Applications
This technology could be used in various fields such as:
- Video editing
- Augmented reality
- Virtual reality
Problems Solved
This technology helps in:
- Efficiently generating videos with specific objects
- Automating the process of object identification in images
Benefits
The benefits of this technology include:
- Saving time and effort in video creation
- Enhancing the quality of videos by including specific objects
Potential Commercial Applications
This technology could be applied in industries such as:
- Entertainment
- Marketing
- Education
Possible Prior Art
One possible prior art could be the use of neural networks for object recognition in images, but the specific application of generating videos using identified objects may be a novel concept.
Unanswered Questions
How does this technology handle complex backgrounds in images during object identification?
The patent application does not provide details on how the technology deals with complex backgrounds when identifying objects in images. This could be crucial in ensuring accurate object recognition in various settings.
What is the computational cost associated with using neural networks for object identification in multiple images?
The patent application does not mention the computational resources required for implementing the neural networks to identify objects in multiple images. Understanding the computational cost could be essential for assessing the feasibility of this technology in practical applications.
Original Abstract Submitted
Apparatuses, systems, and techniques to generate a video using two or more images comprising objects to be included in the video. In at least one embodiment, objects are identified in two or more images using one or more neural networks, to generate a video to include the objects in the video.
