DAVIS346

[Kaleb Nelson]

I got a message from iniVation that the DAVIS346 is now available for purchase at the former DAVIS240C price point. I'm excited for the improvements, but I'm worried that jAER and cAER are not as mature in handling the DAVIS346 as they are with the DAVIS240C. Should I not be worried? I see that the jAER AEChip Menu already has DAVIS346 classes available. 

[Luca Longinotti]

Both jAER and cAER have supported this for a long time (late 2015 for our advanced DAVIS prototype systems).

Also the software is written with support for multiple devices in mind, in fact most of the code is the same in the background.

For example libcaer only has CAER_DAVIS as a device to select, and it uses the same code to open the different resolutions that are available.

So no, on the software support part everything is good to go. If you were able to run a 240C, running a 346 is exactly the same, meaning in jAER you select the appropriate chip class, and in cAER you will get a node called 'DAVIS346' with all its hardware-specific configuration instead of it being called 'DAVIS240C'. And of course if you query for the resolution, you get 346x260 back as a result.

[Tobi Delbruck]

Yes, I agree. We been using this IC that we developed in our EU project SEEBETTER for several years now internally ever since we first got it back from fab. The software is almost exactly compatible between 240 and 346. All the biases and event filters work without change (if they are properly written to get the chip size at runtime).  jAER users need to load configuration (biases and other hardware settings) for this IC/board on first usage.

The Davis346 IC is not published by itself, but we used it for several recent IC papers that have some characterization data, but the characterization is not really complete as it was for DVS128 and DAVIS240 IC papers.

Nozaki, Y., and Delbruck, T. (2017). Temperature and Parasitic Photocurrent Effects in Dynamic Vision Sensors. IEEE Transactions on Electron Devices PP, 1–7. doi:10.1109/TED.2017.2717848.
Taverni, G., Moeys, D. P., Li, C., Cavaco, C., Motsnyi, V., Bello, D. S. S., et al. (2018). Front and Back Illuminated Dynamic and Active Pixel Vision Sensors Comparison. IEEE Transactions on Circuits and Systems II: Express Briefs (accepted) 65, 677–681. Available at: https://ieeexplore.ieee.org/document/8334288/?arnumber=8334288&source=authoralert.
Binas, J., Neil, D., Liu, S.-C., and Delbruck, T. (2017). DDD17: End-To-End DAVIS Driving Dataset. in ICML’17 Workshop on Machine Learning for Autonomous Vehicles (MLAV 2017) (Sydney, Australia). Available at: https://openreview.net/forum?id=HkehpKVG-&noteId=HkehpKVG-.

 

Also, there are several recent papers from our collaborators that have used this sensor in jointly funded projects

Zhu, A. Z., Thakur, D., Ozaslan, T., Pfrommer, B., Kumar, V., and Daniilidis, K. (2018). The Multi Vehicle Stereo Event Camera Dataset: An Event Camera Dataset for 3D Perception. IEEE Robotics and Automation Letters, 1–1. doi:10.1109/LRA.2018.2800793.

Zhu, A. Multi Vehicle Stereo Event Camera Dataset. Available at: https://daniilidis-group.github.io/mvsec/ [Accessed February 21, 2018].

Zhu, A. Z., Yuan, L., Chaney, K., and Daniilidis, K. (2018b). EV-FlowNet: Self-Supervised Optical Flow Estimation for Event-based Cameras. arXiv:1802.06898 [cs]. Available at: http://arxiv.org/abs/1802.06898 [Accessed February 21, 2018].

* A. Rosinol Vidal, H.Rebecq, T. Horstschaefer, D. Scaramuzza

Ultimate SLAM? Combining Events, Images, and IMU for Robust Visual SLAM in HDR and High Speed Scenarios

IEEE Robotics and Automation Letters (RA-L), 2018.

* H. Rebecq, T. Horstschaefer, G. Gallego, D. Scaramuzza

EVO: A Geometric Approach to Event-based 6-DOF Parallel Tracking and Mapping in Real-time

IEEE Robotics and Automation Letters (RA-L), 2016.