Depth of deep learning discussed at Vision Stuttgart
Both the award winners at this year's Vision show in Stuttgart – Prophesee for the Vision Award and GrAI Matter Labs for best start-up – were AI devices. In fact, of the 15 young companies pitching for best start-up, 10 mentioned AI.
Neural networks have become a phenomenon and the solutions addressing industrial imaging and the pain points associated with production lines – the lack of pictures of defects and the challenges of annotating data – are multiplying.
Before announcing the Vision Award winner, Martin Wäny of the judging panel picked out a few past winners, including AnaLogic, which won in 2003 for a neural network processing chip. Wäny said: 'Back then it was debatable whether neural networks were suitable or usable for industrial vision and factory automation. Well, if you walk around this show it’s hard not to find a booth that doesn’t mention AI.'
The VDMA and Messe Stuttgart organised a panel session on the topic of deep learning at the show. Dr Dietmar Ley, CEO of Basler, said during the discussion: ‘We’ve made great strides in making deep learning more usable and more practical over the course of the last three years [since the last Vision show].'
He said: 'Three years ago it was difficult to find the right tools and components to do deep learning properly, and at a reasonable cost point. We believe this has greatly changed.'
He went on to note the abundance of processors now available for running neural networks, and not just GPUs, but processors dedicated for deep learning. As Dr Olaf Munkelt, managing director of MVTec Software, said: 'Giants like Intel have built special neural network cores in their hardware. The hope is that this becomes more cost effective for our customers.'
The winner of the start-up award, GrAI Matter Labs, has done just that. It plans to release an AI processor that it says is 20 times more efficient in terms of inferences per second per Watt than a Google Edge TPU or an Nvidia Jetson NX.
The Life-Ready AI chip is a vision inference processor that lowers energy consumption by only processing changes happening in the images rather than every single pixel. In this way it can optimise the compute inside the network.
Both Framos and Adlink have built solutions based on the chip, with Framos showing a 3D imaging platform running Life-Ready AI with a D435e depth camera in Stuttgart.
GrAI Matter Labs is only one of nine AI chip makers in Europe, compared to 32 firms building AI chips in China and 64 in the US, Christian Verbrugge, GrAI Matter Labs' senior director of business development Europe, pointed out during his pitch to the start-up award judges.
The disparity in the levels of activity around using AI for image processing in different regions was picked up on by Ley during the panel discussion. He said that there's big potential behind the technology and it should be considered on its merits. 'What we see, especially in the Chinese market, is that there are a lot of people experimenting with it, and going through the learning curve now... From a regional perspective, Europeans should get into this technology sooner rather than later, in order to make sure they are not left behind. There are engineers in other territories that are jumping on this.'
Ley said the majority of deep learning projects involving machine vision are custom projects at the moment. He said that the expectation is for large IT companies like Amazon Web Services to release toolchains in the future that make deep learning easy to use and that might work for a couple of standard applications, but this is 'some time down the road', he said.
- Start small and start quickly with AI: advice from Automate Forward
- Edge imaging driven by AI, say Embedded World panel
- Petra Thanner and Daniel Soukup, from the Austrian Institute of Technology, discuss ways of working with CNNs when data is scarce
- Michał Czardybon, CEO of Adaptive Vision, on how to handle data and image annotation when working with deep learning
- Vassilis Tsagaris and Dimitris Kastaniotis at Irida Labs say an iterative approach is needed to build a real-world AI vision application on embedded hardware
Donato Montanari, vice president and general manager, machine vision at Zebra Technologies, said a lot of the challenges organisations face are around how to start using deep learning. Zebra Technologies has recently entered the industrial vision space with a range of smart cameras, alongside acquiring software provider Adaptive Vision. Ley also noted that many of Basler's customers don’t have sufficient competence yet when using the technology.
The first suggestion for companies wanting to start using deep learning is to save images, Montanari advised. 'Take images, take a lot of them, store them, learn how to organise them,' he said.
The next step is learning how to annotate the images, which is where a lot of deployments fail, Montanari continued. He said the best way to annotate images is to be on the factory floor together with the quality expert.
Some of the start-ups pitching as part of the start-up award were offering services to help make data annotation easier and more effective, companies like HodooAI that offers a cloud service for data labelling. Services in data annotation are growing because it's such a critical part to get right – Andrew Ng of Landing AI, speaking at the Association for Advancing Automation's Automate Forward show earlier in the year, said that 80 per cent of the time invested in a deep learning vision project is in dealing with data.
Ultimately, as Munkelt said during the Vision panel discussion, there have to be toolchains developed for the entire process, from labelling data to the choice of algorithm, so that the customer can quickly find out what works and what doesn't.
Inference can be run on a CPU, but training a network typically needs a GPU, Montanari said. There's also the option of training a network in the cloud and then deploying it to edge devices.
Stephen Walsh at Neurala Europe, one of the start-ups, noted during his pitch that GPUs are expensive and that they are in short supply at the moment. He said Neurala is aiming to reduce the cost of deploying deep learning and to make it more accessible. The company has partnered with Flir to offer Neurala tools and run a Neurala VIA model on Flir's Firefly DL camera.
Walsh said that it's important to accept the fact that there aren't many images of defects available in industrial inspection, and that 'we have to be able to train with that'.
Jens Hülsmann at Isra Vision said deep learning is 'clearly a future technology'. He said it's another tool in the tool box that can solve tasks that were difficult with classic imaging processing methods.
Mark Williamson at Stemmer Imaging noted during the panel discussion that there is now technology developed to make it easier to train a network with fewer defect images, and that over the last three years deep learning has become easier to deploy with examples in industrial applications.
He said: 'Like any technology, you can be too early, you can be too late, you can be at the right time. Neural networks have been talked about since the 90s. That was too early but there was potential. We’re now getting to the point where we’re seeing adoption in some applications. Certainly it’s got its future. Now is the moment [to adopt].'
Products shown at Vision
- Musashi AI has worked with automotive firms to develop AI inspection for complex parts like camshafts. The software combines anomaly detection and supervised learning with object detection. It has the ability to start inspecting with 30-50 good parts, and use the anomaly detector to correlate what the object detector sees to identify defects. It all runs on the edge. Version 2.0 also uses instance segmentation to improve the process.
- EVT was showing a smart 3D triangulation camera built on a Google Coral chip for running deep learning, either for pre-processing or analysing the point cloud. The camera also contains a Xilinx Zynq dual core processor with FPGA.
The Coral SoC for deep learning operates at 4TOPS and 4W TDP power consumption. Those developing in TensorFlow can convert their networks to TensorFlow Lite and run it onboard the camera in the Coral chip. Transfer learning can also be done on the smart camera, meaning the network can be trained on-site.
Other EVT cameras are also available based on similar processing hardware, including a dual head camera, a line scan model and a thermal camera.
- Easics has developed an embedded neural network technology called NearbAI. The technology is designed for running neural networks close to the sensor on embedded hardware, typically in an application with low latency. NearbAI is platform independent and can be mapped to FPGAs, ASICs, and SoMs. It can be used for tasks requiring ultra-low latency as well as non-variable latency.