Area Efficient Compression for Floating-Point Feature Maps in Convolutional Neural Network Accelerators.
Bai-Kui YanShanq-Jang RuanPublished in: IEEE Trans. Circuits Syst. II Express Briefs (2023)
Keyphrases
- floating point
- feature maps
- convolutional neural network
- efficient compression
- bit rate
- higher quality
- neural network
- face detection
- video compression
- image compression
- self organizing maps
- kernel function
- video coding
- video codec
- fixed point
- saliency map
- feature vectors
- feature space
- floating point arithmetic
- image quality
- fuzzy neural network
- wavelet transform
- motion estimation
- visual quality
- instruction set
- high quality
- motion vectors
- low level
- computational complexity
- face recognition
- computer vision