Enhancing the Energy Efficiency and Robustness of tinyML Computer Vision Using Coarsely-quantized Log-gradient Input Images.
Qianyun LuBoris MurmannPublished in: ACM Trans. Embed. Comput. Syst. (2024)
Keyphrases
- input image
- energy efficiency
- computer vision
- power consumption
- energy consumption
- wireless sensor networks
- image gradient
- sensor networks
- data center
- energy efficient
- response time
- routing protocol
- energy saving
- energy conservation
- smart home
- depth map
- power saving
- power management
- image structure
- image processing
- image pairs
- computer graphics
- object recognition
- image sequences
- wireless sensor
- computer vision algorithms
- specular reflection
- action recognition
- image matching
- color images
- optical flow
- image segmentation
- artificial intelligence
- multi hop
- residual energy
- data sets
- routing algorithm
- sensor data
- context aware
- input data
- motion estimation
- data streams
- face recognition