# Energy-Efficient Visual Inspection with FFT-Based CNNs and Adaptive Floating-Point Quantization

arXiv: https://arxiv.org/abs/2608.19837 · HTML: https://mlsanity.com/paper/2608.19837/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 2 of its 7 references that are in the map.
- First posted on arXiv 20 Aug 2026.

## Builds on (most cited first, 2 of 2)

- [FP8 Formats for Deep Learning](https://mlsanity.com/paper/2209.05433/) (2022)
- [How to keep pushing ML accelerator performance? Know your rooflines!](https://mlsanity.com/paper/2505.16346/) (2025)

## Abstract (arXiv)

This paper investigates reduced-precision floating-point arithmetic for FFT-based CNN inference on an industrial CPU-FPGA platform. We combine FFT-based convolution with adaptive post-training FP8 quantization and evaluate two FPGA-oriented optimization methods: progressive bias adjustment (PBA) within the FFT and layer-wise exponent-bias selection across the CNN. The methods are implemented in a LeNet-5 accelerator using serial radix-22 SDF FFT modules and evaluated on an industrial fault detection dataset. Results show that weight scaling outperforms PBA, while layer-wise bias optimization increases the accuracy from 80.33% to 84.13% without modifying the datapath width. Compared with CPU-only inference, the FPGA achieves approximately 2.5× higher energy efficiency.