18242965. GENERATING IMAGES WITH SMALL OBJECTS FOR TRAINING A PRUNED SUPER-RESOLUTION NETWORK simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
Contents
- 1 GENERATING IMAGES WITH SMALL OBJECTS FOR TRAINING A PRUNED SUPER-RESOLUTION NETWORK
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 GENERATING IMAGES WITH SMALL OBJECTS FOR TRAINING A PRUNED SUPER-RESOLUTION NETWORK - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
GENERATING IMAGES WITH SMALL OBJECTS FOR TRAINING A PRUNED SUPER-RESOLUTION NETWORK
Organization Name
Inventor(s)
Kamal Jnawali of Tustin CA (US)
Tien Cheng Bau of Irvine CA (US)
GENERATING IMAGES WITH SMALL OBJECTS FOR TRAINING A PRUNED SUPER-RESOLUTION NETWORK - A simplified explanation of the abstract
This abstract first appeared for US patent application 18242965 titled 'GENERATING IMAGES WITH SMALL OBJECTS FOR TRAINING A PRUNED SUPER-RESOLUTION NETWORK
Simplified Explanation
The abstract describes a method for detecting objects in input video frames, cropping images containing these objects, overlaying simulated text on the cropped images, and providing them to a pruned convolutional neural network for learning to reconstruct objects and textual regions during image super-resolution.
- Detect objects in input video frames
- Crop images containing detected objects
- Overlay simulated text on cropped images
- Provide training images to a pruned CNN
- CNN learns to reconstruct objects and textual regions during image super-resolution
Potential Applications
This technology could be applied in various fields such as video surveillance, image enhancement, and object recognition systems.
Problems Solved
This technology helps in improving the quality of images by reconstructing objects and textual regions during image super-resolution, which can be beneficial in enhancing visual content.
Benefits
The benefits of this technology include improved image quality, better object recognition, and enhanced visual content in videos.
Potential Commercial Applications
A potential commercial application of this technology could be in the development of advanced video surveillance systems for security purposes.
Possible Prior Art
One possible prior art for this technology could be existing image super-resolution techniques that focus on enhancing image quality through various methods.
What are the specific techniques used for object detection in input video frames?
The abstract does not provide specific details on the techniques used for object detection in input video frames.
How does the pruned convolutional neural network learn to reconstruct objects and textual regions during image super-resolution?
The abstract does not elaborate on the specific learning process of the pruned convolutional neural network for reconstructing objects and textual regions during image super-resolution.
Original Abstract Submitted
One embodiment provides a method comprising detecting at least one object displayed within at least one input frame of an input video. The method further comprises cropping, from the at least one input frame, at least one cropped image including the at least one object. The method further comprises generating at least one training image by overlaying simulated text on the at least one cropped image. The method further comprises providing the at least one training image to a pruned convolutional neural network (CNN). The pruned CNN learns, from the at least one training image, to reconstruct objects and textual regions during image super-resolution.
