Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.

Survey: Feedback and Improvements

idata
Employee
1,445 Views

Please feel free to add any comments by replying to this thread.

0 Kudos
22 Replies
idata
Employee
1,103 Views

I hope to see support for Mask R-CNN very soon.

0 Kudos
idata
Employee
1,103 Views

@GoldenWings Thanks for the input. Any other suggestions or features you guys would want to see?

0 Kudos
idata
Employee
1,103 Views

Api corresponding to distributed processing to multiple sticks is necessary, not cyclic processing.

0 Kudos
idata
Employee
1,103 Views

Would like to see support for squeezenet and Tensorflow mobilenet SSD object detection.

0 Kudos
idata
Employee
1,103 Views

I would in general like to see more support for different tensorflow ops. This would enable greater freedom in design of networks which would be really cool. :) Also more tutorials for maybe custom Networks. :)

0 Kudos
idata
Employee
1,103 Views

Support for Tensorflow mobilenet SSD. The blog posts at movidius.github.io are quite good - additional topics by the same authors would be helpful.

0 Kudos
idata
Employee
1,103 Views

Tutorials on getting new networks not already in the ncappzoo working on the sticks. Like DSOD - Densely Supervised Object Detectors. MobileNets-SSD is slow to train on Caffe. (and will remain until they start supporting an efficient Depthwise Separable Layer, in Caffe-SSD)

0 Kudos
idata
Employee
1,103 Views

@trevor you should be able to use this link for training SSD with efficient DWS convolution with CUDNN9: https://github.com/listenlink/caffe/tree/ssd

0 Kudos
idata
Employee
1,103 Views

Please support Keras. I have everything setup with Keras in Ubuntu. I do not see particular reason to add Caffe in my system other than Movidius stick. Untill then example graph files that can be deployed in RPi would be appreciated.

0 Kudos
idata
Employee
1,103 Views

How to use retrain.py and your own classifications, mnvcompile and use it on the stick.

0 Kudos
idata
Employee
1,103 Views

     

  1. Yolo support in all flavors ( not just tiny .. I prefer correct classification above fast&wrong )

     

    I would like to try Yolo v2 and v3…
  2.  

  3. tested and forgiving installation script ( It's not fun to install NCSDK2 under Ubuntu studio/Mate 16.04/18.04 )
  4.  

  5. Update all samples from appzoo to run under version 2
  6.  

  7. Movidius "under the hood", low level api ?
  8.  

 

Release date for Movidius X ??? ( this year !? )

0 Kudos
idata
Employee
1,103 Views

It would be great if the stick will support Pytorch in the future

0 Kudos
idata
Employee
1,103 Views

1) NCS kit in general is messy and loaded with Python dependencies. Straight C/C++ solutions would be best and a little python maybe for conversion / support.

 

2) install.sh just fails in so many ways on a vanilla clean install Ubuntu 16.04 - like it was never really tested.

 

3) find a way to enable a broader support community by open sourcing more tools (e.g. myriad compilers) and selecting a few outside partners to support smaller hobbyist / maker types.

 

4) get a partner program going - important to all of the above items
0 Kudos
idata
Employee
1,103 Views
1 request is MobileNet SSD for TensorFlow.

 

It would also be nice for better Raspberry Pi / TinkerBoard support during installation and compilation.

0 Kudos
idata
Employee
1,103 Views

Internal processing FP32 compatible

0 Kudos
idata
Employee
1,103 Views

     

  • Correspondence to FusedBatchNormV2
  •  

  • Correspondence to Caffe-Segnet
  •  

0 Kudos
idata
Employee
1,103 Views

     

  • Correspondence to caffe-jacinto
  •  

0 Kudos
idata
Employee
1,103 Views

@Tome_at_Intel please merge outstanding pull requests and support the Github code! I just made a PR to fix a simple syntax bug, and I now see someone made the exact same pull request on the 27th Feb, 8 MONTHS ago, and it hasn't been merged!

 

https://github.com/movidius/ncsdk/pulls

0 Kudos
idata
Employee
1,103 Views

I'm working on face recognition using MTCNN and Sphereface, and found that

 

     

  1. both v1 and v2 cannot change the input shape/ cannot get multi-output *(and when I try to concatenate the outputs.. bugs come, even worse is that the outputs can't be concatenate..)
  2.  

  3. v2 doesn't support MTCNN, there's always info saying that no weights provided, list index out of range when I try to compile the graph. (btw, somehow many warning messages aren't that specific, they are even misleading)

     

    (2.1 No demo for loading multi-graph in one ncs which should be a huge achievement in v2(sorry that I don't know the fifo is bound to the graph or the device? Could you give more details about the fifo?))

     

    (2.2 plz try to support more and more layers that is normal in tf or caffe afap..thx.. orz.. )
  4.  

0 Kudos
idata
Employee
856 Views

Multioutput Support many awesome tensorflow models has many outputs, and the only model that the documentation teach you to compile to graph is incepction v3

0 Kudos
Reply