Intel corporation (20240338789). CONTROL OF INPUT DIMENSIONS FOR COMPUTER VISION MODEL TRAINING simplified abstract
Contents
CONTROL OF INPUT DIMENSIONS FOR COMPUTER VISION MODEL TRAINING
Organization Name
Inventor(s)
CONTROL OF INPUT DIMENSIONS FOR COMPUTER VISION MODEL TRAINING - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240338789 titled 'CONTROL OF INPUT DIMENSIONS FOR COMPUTER VISION MODEL TRAINING
The patent application describes an apparatus that determines the optimal spatial input size for training a computer vision model based on the sizes of input training images and objects within them.
- The apparatus adjusts the initial spatial input size to account for object sizes in the training images and maps it to an available set of spatial input sizes.
- It evaluates a linear model to determine the final batch size for training the computer vision model, based on simulations using different spatial input sizes and batch sizes.
- The innovation aims to optimize the training process of computer vision models by dynamically adjusting input sizes and batch sizes for improved performance.
Potential Applications:
- This technology can be applied in various industries such as autonomous vehicles, healthcare imaging, and security surveillance systems.
- It can enhance the accuracy and efficiency of object detection, image classification, and other computer vision tasks.
Problems Solved:
- Addresses the challenge of determining the optimal input size and batch size for training computer vision models.
- Improves the performance and accuracy of computer vision systems by dynamically adjusting input parameters.
Benefits:
- Enhanced accuracy and efficiency in computer vision tasks.
- Improved performance of computer vision models through optimized input sizes and batch sizes.
Commercial Applications:
- This technology has commercial applications in industries such as autonomous driving, medical imaging, and video surveillance for enhanced object detection and image analysis capabilities.
Questions about the Technology: 1. How does the apparatus determine the adjusted spatial input size based on object sizes in training images? 2. What are the potential implications of using a linear model to determine the final batch size for training computer vision models?
Original Abstract Submitted
example apparatus disclosed herein determine an initial spatial input size for training a computer vision model, the initial spatial input size based on sizes of input training images, apply an adjustment to the initial spatial input size to determine an adjusted spatial input size, the adjustment based on sizes of objects in the input training images, and map the adjusted spatial input size to one of a set of available spatial input sizes to determine a final spatial input size for training the computer vision model. some disclosed apparatus evaluates a linear model to determine a final batch size for training the computer vision model, the linear model based on first and second simulations of training the computer vision model, the first simulation based on the final spatial input size and a first batch size, the second simulation based on the final spatial input size and a second batch size.
