AMD OpenCL-OpenGL interoperability on Ubuntu Linux
As far as I know, the OpenCL-OpenGL interopability on Linux only works with the amdgpu pro driver. In this post I explain how you can get the CL-GL interop to work on ubuntu. Start...
Erik Smistad Blog
Deep learning inference engines
I have been working a lot lately with different deep learning inference engines, integrating them into the FAST framework. Specifically I have been working with Google’s TensorFlow (with cuDNN acceleration), NVIDIA’s TensorRT and Intel’s...
Highlighting nerves and blood vessels for ultrasound axillary nerve blocks using neural networks
Just published an article in Journal of Medical Imaging on using neural networks to highlight nerves and blood vessels seen during ultrasound-guided axillary nerve blocks. The video above shows the neural network in action...
Norges fylker og kommuner i GeoJSON format
Her finner du Norges fylke og kommunegrenser i GeoJSON format og EPSG:3857 projeksjon. Disse grensene er opprinnelig hentet fra Geonorge, Norges offisielle kart database. Grensekoordinatene fra Geonorge er bare tilgjengelig i UTM 33 projeksjon,...
Using deep neural networks to automate cardiac measurements in 2D ultrasound in real-time
This video shows an application we are currently working on at the Centre for Innovative Ultrasound Solutions (CIUS). The application is developed using FAST which is integrated with TensorFlow for fast neural network inference.
Fix NVIDIA driver 387 for Linux kernel 4.13+ on Ubuntu
Apparently, the current NVIDIA drivers (387) which comes with CUDA 9.1 doesn’t compile with the new Linux kernels (4.13+) on Ubuntu. The error report from DKMS will say something like: /var/lib/dkms/nvidia-387/387.26/build/nvidia-uvm/uvm8_va_block.c: In function ‘block_cpu_fault_locked’:...
Object tracking with Kinect and FAST
This video shows the work in progress on an algorithm for tracking objects captured by a Kinect camera using the FAST framework. FAST now comes with a Kinect module which uses libfreenect2 to stream...
Vessel detection in ultrasound images using deep convolutional neural networks
The video shows our first result in the field of deep learning and convolutional neural networks. We designed a neural network which can automatically locate and segment blood vessels in real-time from B-mode ultrasound...
Visualizing learned features of a caffe neural network
After training a convolutional neural network, one often wants to see what the network has learned. The following python function creates and displays an image with all convolutions of a specific layer as shown...
OpenCL bugs
OpenCL is a great framework, however, AMD’s, NVIDIA’s and Apple’s OpenCL compilers are not. I have worked over 4 years with OpenCL and I have experienced a lot of bugs. Here I present a...
