Qualcomm incorporated (20240161368). REGENERATIVE LEARNING TO ENHANCE DENSE PREDICTION simplified abstract
Contents
- 1 REGENERATIVE LEARNING TO ENHANCE DENSE PREDICTION
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 REGENERATIVE LEARNING TO ENHANCE DENSE PREDICTION - 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 Unanswered Questions
- 1.11 Original Abstract Submitted
REGENERATIVE LEARNING TO ENHANCE DENSE PREDICTION
Organization Name
Inventor(s)
Shubhankar Mangesh Borse of San Diego CA (US)
Debasmit Das of San Diego CA (US)
Hyojin Park of San Diego CA (US)
Risheek Garrepalli of San Diego CA (US)
Fatih Murat Porikli of San Diego CA (US)
REGENERATIVE LEARNING TO ENHANCE DENSE PREDICTION - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240161368 titled 'REGENERATIVE LEARNING TO ENHANCE DENSE PREDICTION
Simplified Explanation
The present disclosure describes techniques for regenerative learning to enhance dense predictions using machine learning models.
- Input image is accessed
- Dense prediction output is generated based on the input image using a machine learning model
- Regenerated version of the input image is generated
- First loss is generated based on the input image and corresponding ground truth dense prediction
- Second loss is generated based on the regenerated version of the input image
- Parameters of the machine learning model are updated based on the losses
Potential Applications
This technology could be applied in various fields such as image recognition, medical imaging, autonomous driving, and video analysis.
Problems Solved
This technology helps improve the accuracy and reliability of dense predictions in machine learning models.
Benefits
Enhanced dense predictions Improved performance of machine learning models Increased efficiency in image processing tasks
Potential Commercial Applications
Potential commercial applications include image recognition software, medical imaging systems, autonomous vehicles, and video surveillance systems.
Possible Prior Art
Prior art in the field of machine learning and image processing may include techniques for improving prediction accuracy and model performance through loss functions and parameter updates.
Unanswered Questions
How does this technology compare to other methods for enhancing dense predictions in machine learning models?
This article does not provide a direct comparison with other methods or technologies in the field.
What are the specific industries or sectors that could benefit the most from this technology?
The article does not specify which industries or sectors could benefit the most from the described technology.
Original Abstract Submitted
certain aspects of the present disclosure provide techniques and apparatus for regenerative learning to enhance dense predictions. in one example method, an input image is accessed. a dense prediction output is generated based on the input image using a dense prediction machine learning (ml) model, and a regenerated version of the input image is generated. a first loss is generated based on the input image and a corresponding ground truth dense prediction, and a second loss is generated based on the regenerated version of the input image. one or more parameters of the dense prediction ml model are updated based on the first and second losses.
