Alert button
Picture for Christoph Gerum

Christoph Gerum

Alert button

Precise localization within the GI tract by combining classification of CNNs and time-series analysis of HMMs

Oct 11, 2023
Julia Werner, Christoph Gerum, Moritz Reiber, Jörg Nick, Oliver Bringmann

Figure 1 for Precise localization within the GI tract by combining classification of CNNs and time-series analysis of HMMs
Figure 2 for Precise localization within the GI tract by combining classification of CNNs and time-series analysis of HMMs
Figure 3 for Precise localization within the GI tract by combining classification of CNNs and time-series analysis of HMMs
Figure 4 for Precise localization within the GI tract by combining classification of CNNs and time-series analysis of HMMs

This paper presents a method to efficiently classify the gastroenterologic section of images derived from Video Capsule Endoscopy (VCE) studies by exploring the combination of a Convolutional Neural Network (CNN) for classification with the time-series analysis properties of a Hidden Markov Model (HMM). It is demonstrated that successive time-series analysis identifies and corrects errors in the CNN output. Our approach achieves an accuracy of $98.04\%$ on the Rhode Island (RI) Gastroenterology dataset. This allows for precise localization within the gastrointestinal (GI) tract while requiring only approximately 1M parameters and thus, provides a method suitable for low power devices

* Accepted at MLMI 2023 
Viaarxiv icon

Hardware Accelerator and Neural Network Co-Optimization for Ultra-Low-Power Audio Processing Devices

Sep 08, 2022
Christoph Gerum, Adrian Frischknecht, Tobias Hald, Paul Palomero Bernardo, Konstantin Lübeck, Olver Bringmann

Figure 1 for Hardware Accelerator and Neural Network Co-Optimization for Ultra-Low-Power Audio Processing Devices
Figure 2 for Hardware Accelerator and Neural Network Co-Optimization for Ultra-Low-Power Audio Processing Devices
Figure 3 for Hardware Accelerator and Neural Network Co-Optimization for Ultra-Low-Power Audio Processing Devices
Figure 4 for Hardware Accelerator and Neural Network Co-Optimization for Ultra-Low-Power Audio Processing Devices

The increasing spread of artificial neural networks does not stop at ultralow-power edge devices. However, these very often have high computational demand and require specialized hardware accelerators to ensure the design meets power and performance constraints. The manual optimization of neural networks along with the corresponding hardware accelerators can be very challenging. This paper presents HANNAH (Hardware Accelerator and Neural Network seArcH), a framework for automated and combined hardware/software co-design of deep neural networks and hardware accelerators for resource and power-constrained edge devices. The optimization approach uses an evolution-based search algorithm, a neural network template technique, and analytical KPI models for the configurable UltraTrail hardware accelerator template to find an optimized neural network and accelerator configuration. We demonstrate that HANNAH can find suitable neural networks with minimized power consumption and high accuracy for different audio classification tasks such as single-class wake word detection, multi-class keyword detection, and voice activity detection, which are superior to the related work.

* Accepted Version for: EUROMICRO DSD 2022 
Viaarxiv icon

Behavior of Keyword Spotting Networks Under Noisy Conditions

Sep 15, 2021
Anwesh Mohanty, Adrian Frischknecht, Christoph Gerum, Oliver Bringmann

Figure 1 for Behavior of Keyword Spotting Networks Under Noisy Conditions
Figure 2 for Behavior of Keyword Spotting Networks Under Noisy Conditions
Figure 3 for Behavior of Keyword Spotting Networks Under Noisy Conditions
Figure 4 for Behavior of Keyword Spotting Networks Under Noisy Conditions

Keyword spotting (KWS) is becoming a ubiquitous need with the advancement in artificial intelligence and smart devices. Recent work in this field have focused on several different architectures to achieve good results on datasets with low to moderate noise. However, the performance of these models deteriorates under high noise conditions as shown by our experiments. In our paper, we present an extensive comparison between state-of-the-art KWS networks under various noisy conditions. We also suggest adaptive batch normalization as a technique to improve the performance of the networks when the noise files are unknown during the training phase. The results of such high noise characterization enable future work in developing models that perform better in the aforementioned conditions.

* ICANN 2021. Lecture Notes in Computer Science, vol 12891, pp 369-378. Springer  
* 11 pages, 5 figures, Published in Lecture Notes in Computer Science book series (LNCS, volume 12891) 
Viaarxiv icon