profile
viewpoint
Maksim Surguy msurguy @amazon Seattle, WA http://maxoffsky.com MS in Tech Innovation, Design Technologist at Amazon Past: dev @firstech, @nbcuniversal, creator of Bootsnipp and many more. On Twitter: @msurguy

msurguy/background-blur 982

Ultra light cross browser image blurring plugin for jQuery

msurguy/awesome-3d-engines-for-plotters 40

Collect links to awesome 3D SVG exporters

msurguy/awesome-webxr 34

Compilation of links for WebXR

msurguy/awesome-jetson-nano 7

Highlight resources and tools for Nvidia Jetson Nano

msurguy/awesome-machine-learning 5

A curated list of awesome Machine Learning frameworks, libraries and software.

msurguy/arduino-wiichuck-laser 3

Arduino Wiichuck driven servos with a Laser

msurguy/awesome-plotters 3

A curated list of code and resources for pen plotters and other robots that draw

adamgoose/laravel-docs 2

A Crowd-Sourced Laravel Documentation

ManiacalLabs/BiblioPixelNeoSegment 2

Adds BiblioPixel support for the NeoSegment: https://www.crowdsupply.com/maksmakes/neosegment

msurguy/all-scad-on-github 2

List of all OpenSCAD files on github

starteddferber90/playroom-clickable-prototypes

started time in an hour

issue openeddusty-nv/jetson-inference

[TRT] Could not register plugin creator - ::FlattenConcat_TRT version 1

I am using the video tutorial on Youtube to develop my first real time object detection. When I try to run the script, the following error show up [TRT] Could not register plugin creator - ::FlattenConcat_TRT version 1 After the error, the camera never turns on and the terminal keeps on spitting values like this: After concat removal: 66 layers [TRT] Graph construction and optimization completed in 0.045886 seconds. [TRT] Constructing optimization profile number 0 [1/1]. [TRT] --------------- Timing Runner: <reformat> (Reformat) [TRT] Tactic: 1002 time 0.747864 [TRT] Tactic: 0 time 1.23659 [TRT] Fastest Tactic: 1002 Time: 0.747864 [TRT] --------------- Timing Runner: <reformat> (Reformat) [TRT] Tactic: 1002 time 20.0891 [TRT] Tactic: 0 time 1.31471 [TRT] Fastest Tactic: 0 Time: 1.31471

Has anyone come up with the same issue and knows how to fix it? I'm running the script on the Jetson Nano 4GB Development kit with Jetpack 4.4 installed and using the Raspberry Pi Camera V2.

This is the actual script:

import jetson.utils

net = jetson.inference.detectNet("coco-dog", threshold=0.5)
camera = jetson.utils.videoSource("csi://0")      # '/dev/video0' for V4L2
display = jetson.utils.videoOutput("display://0") # 'my_video.mp4' for file

while display.IsStreaming():
        img = camera.Capture()
        detections = net.Detect(img)
        display.Render(img)
        display.SetStatus("Object Detection | Network {:.0f} FPS".format(net.GetNetworkFPS()))```

created time in 2 hours

push eventChristian-Nunnally/Diiagramr

Christian Nunnally

commit sha 93a41f766a0aebcebf295776cc25195f5b0aa5a8

Setting startup project

view details

push time in 2 hours

push eventChristian-Nunnally/Diiagramr

Christian Nunnally

commit sha 810b0a670f092befae467a67f6a9a9ae97419e9e

Made project dirtying work. Cleaned up some todos. Fixed a bug where opening a new project with a dirty project open would close the new project as well as the current one when the dialog was closed.

view details

Christian Nunnally

commit sha 5c0c48e3a4f08dd6ffb75b84c9f6c4844e445375

Making first time setup easier.

view details

push time in 3 hours

startedbarrust/pyprobables

started time in 3 hours

startedturu/yalal

started time in 3 hours

issue commentdusty-nv/jetson-inference

External Power Supply for Jetson Nano

@Alex-Beh Anything with 5V and 4 Amps will work. The ampere is the problem since most voltage at that level wont supply that much current on a portable device. It turns out that you can set the max current but that will affect the performance. Long story short I bought "Krisdonia Portable Laptop Charger (TSA-Approved) 92.5Wh/25000mAh" and it works. However, I recommend you look into batteries more along the lines of those specification

achen173

comment created time in 4 hours

startedbitnami/bitnami-docker-pgpool

started time in 4 hours

issue commentgeerlingguy/pi-webcam

Feature request: mirror camera output to HDMI port.

I guess you could do this with a script that runs VLC on boot? https://www.raspberrypi.org/forums/viewtopic.php?t=80518

iampegram

comment created time in 4 hours

issue commentgeerlingguy/pi-webcam

Not working on Raspberry Pi 4B or A/A+

Using the old kernal work around on 4b, camera only works in Facetime on Mac, no other apps recognize the can. Works fine on PC. Interestingly using 4b alpha version of show me webcam only works on Mac if cam is opened first in Photo Booth, then extremely laggy. Scrambled vertical bars otherwise.

geerlingguy

comment created time in 4 hours

issue openedgeerlingguy/pi-webcam

Suggestion: Use gcc optimizations for building uvc-gadget

I tried building uvc-gadget with this configuration in Makefile:

CFLAGS          := -W -Wall -O3 $(KERNEL_INCLUDE)
LDFLAGS         := -O3

I don't see any dropped frames. CPU utilization is also pretty low.

created time in 4 hours

startedinnoave/genevo

started time in 5 hours

startedxsrf/pcb3-avr

started time in 5 hours

issue openedgeerlingguy/pi-webcam

Feature request: mirror camera output to HDMI port.

It would be nice to be able to use this as an HDMI camera as well.

created time in 6 hours

fork schollz/hachi

808 drum machine for monome norns

fork in 7 hours

startedpangrus/hachi

started time in 7 hours

startedsubprotocol/genetic-js

started time in 7 hours

fork schollz/norns

norns is many sound instruments.

http://monome.org

fork in 7 hours

PR opened tamasmeszaros/libnest2d

CMake: Removing hardcoded install locations

CMake is handling these paths quite well already. When creating my package here it causes also extra work to place the files into the correct locations. For example, when working with debhelper to build the packages for Debian, I normally need to set CMAKE_LIBRARY_PATH=$(DEB_HOST_MULTIARCH) and the libs will be placed into "/usr/lib/x86_64-linux-gnu/" perfectly fine. Removing the hardcoded locations here, will recover the functionality.

+1 -3

0 comment

1 changed file

pr created time in 10 hours

PR opened tamasmeszaros/libnest2d

README: Adding libnlopt-cxx-devw

Tried to build the library locally here and noticed that Ubuntu Focal has a separate package for the C++ library. For this reason, libnlopt-cxx-dev is needed here. Mentioning libnlopt-dev is not needed here since libnlopt-cxx-dev depends on it.

+1 -1

0 comment

1 changed file

pr created time in 11 hours

startedJoeyDeVries/LearnOpenGL

started time in 11 hours

startedNatronGitHub/natron-plugins

started time in 12 hours

startedkfarr/streetmix3d

started time in 15 hours

push eventXXIIVV/oscean

neauoire

commit sha 2d224d7111468136f678b14b7b7ac149a26f84ac

*

view details

push time in 16 hours

issue commentJialeCao001/SipMask

train on a custom dataset

@Traderain For image instance segmentation, you can use coco-style annotations with SipMask-benchmark or SipMask-mmdet. For video instance segmentation, Sipmask-VIS only supports the yotube-vis style annotations.

amaanda

comment created time in 17 hours

issue closeddusty-nv/jetson-inference

error: MESA-LOADER: failed to open swrast

Hi, Thank you for making this great tutorial! I have installed the project but when I try to run the imagenet-camera.py demo, I got this error

Lark20201204-144057

And show no picture as below when the program is finished. It seems that my OpenGL lin is broken? Have you ever encountered this issue? Lark20201204-173617

Thank you!

closed time in 17 hours

hejun888

issue commentdusty-nv/jetson-inference

error: MESA-LOADER: failed to open swrast

thanks,now, it's OK

hejun888

comment created time in 17 hours

startedMagicStack/immutables

started time in 19 hours

issue commentdusty-nv/jetson-inference

Image manipulation with Cuda - questions

Hi @dusty-nv ,

That makes sense. Is there considerable overhead to do the CV -> CUDA -> CV conversion if I have image manipulation functions?

Also with the saving image function, is it faster than saving with cv2?

This topic is quite fascinating !

marvision-ai

comment created time in 20 hours

startedcli-guidelines/cli-guidelines

started time in 21 hours

more