# Jetson Nano best person detector model **URL:** \ **Category:** Jetson Nano\ **Tags:** jetson-inference, tensorrt\ **Created:** [March 31, 2021, 1:09pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845 "2021-03-31T13:09:55Z")\ **Posts on this page:** 11\ **Page:** 1
**Author:** ![codeforge](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@codeforge](https://forums.developer.nvidia.com/u/codeforge)\ **Post date:** [March 31, 2021, 1:09pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/1 "2021-03-31T13:09:55Z")
Hi, I see that there are many example on object detection for Nvidia platform, so I don’t know where to start. I would to use my Jetson Nano to detect and track only person from a webcam. Could you suggest me a models? I see that with ssd-mobilenet-v2 I can get good fps. There is a way to select just person using detectnet in Jetson-inference or I need to re-train ssd-mobilenet-v1 only for person? Thanks ---
**Author:** ![mehmetdeniz](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/mehmetdeniz/32/72318_2.png) [@mehmetdeniz](https://forums.developer.nvidia.com/u/mehmetdeniz)\ **Post date:** [March 31, 2021, 1:44pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/2 "2021-03-31T13:44:42Z")
> [@codeforge](#): > > I would to use my Jetson Nano to detect and track only person from a webcam. If you want to find only people, you may train. Otherwise, you should filter labels in your script (close other detected objects’ labels). ---
**Author:** ![ozguryildiz](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/ozguryildiz/32/72318_2.png) [@ozguryildiz](https://forums.developer.nvidia.com/u/ozguryildiz)\ **Post date:** [March 31, 2021, 1:59pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/3 "2021-03-31T13:59:42Z")
Hi, There is a post below which emphasizes speed and accuracy between different models. [https://jonathan-hui.medium.com/object-detection-speed-and-accuracy-comparison-faster-r-cnn-r-fcn-ssd-and-yolo-5425656ae359](https://jonathan-hui.medium.com/object-detection-speed-and-accuracy-comparison-faster-r-cnn-r-fcn-ssd-and-yolo-5425656ae359) Yolo seems to be fast and more accurant when you compare the models. You can train you custom Yolo model. You can check this link for training custom model. > **[How to Detect Objects in Real-Time Using OpenCV and Python | Towards Data...](https://towardsdatascience.com/how-to-detect-objects-in-real-time-using-opencv-and-python-c1ba0c2c69c0/)** > > For the uninitiated, Real-Time Object Detection might sound quite a mouthful. However, with a few awesome libraries at hand, the job... ---
**Author:** ![dusty\_nv](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/dusty_nv/32/14043_2.png) [@dusty\_nv](https://forums.developer.nvidia.com/u/dusty_nv)\ **Post date:** [March 31, 2021, 5:05pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/4 "2021-03-31T17:05:35Z")
You can check out the production-quality PeopleNet model here: [https://ngc.nvidia.com/catalog/models/nvidia:tlt\_peoplenet](https://ngc.nvidia.com/catalog/models/nvidia:tlt_peoplenet) It was trained/pruned using Transfer Learning Toolkit and you can run it with DeepStream for the best performance. ---
**Author:** ![codeforge](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@codeforge](https://forums.developer.nvidia.com/u/codeforge)\ **Post date:** [April 1, 2021, 10:06am UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/5 "2021-04-01T10:06:50Z")
Thank you so much for fast reply :) , another question, I’m trying to crop a cudaImage from jetson-inference example (detectnet.py) to get a 360 degree image from insta360 oneX2. This is what I get from /dev/video1: ![](https://global.discourse-cdn.com/nvidia/original/3X/7/4/7408d8f1815ab9c28d005334e3be39d495d52f3c.gif) I need to crop the image half height and then concatenate the image horizontally. Using cv2 and numpy I can do it simply but I don’t know how to do using cudaCrop. I can successfully crop a cudaImage but I don’t know how to concatenate the 2 images horizontally. I try to convert it to NumPy and then back to cudaImage and give it to detections but I got many errors .Is there a simple way to modify the input source for detectnet.py? ---
**Author:** ![dusty\_nv](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/dusty_nv/32/14043_2.png) [@dusty\_nv](https://forums.developer.nvidia.com/u/dusty_nv)\ **Post date:** [April 1, 2021, 2:26pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/6 "2021-04-01T14:26:43Z")
Hi @codeforge, you can use the `cudaOverlay()` function to compost two images into an output image. See here for example - [https://github.com/dusty-nv/jetson-inference/blob/master/docs/aux-image.md#overlay](https://github.com/dusty-nv/jetson-inference/blob/master/docs/aux-image.md#overlay) ---
**Author:** ![codeforge](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@codeforge](https://forums.developer.nvidia.com/u/codeforge)\ **Post date:** [April 2, 2021, 9:12am UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/7 "2021-04-02T09:12:40Z")
I will try that, thank u. ---
**Author:** ![loophole64](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/loophole64/32/38308_2.png) [@loophole64](https://forums.developer.nvidia.com/u/loophole64)\ **Post date:** [May 10, 2021, 8:00pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/8 "2021-05-10T20:00:44Z")
> [@dusty\_nv](#): > > You can check out the production-quality PeopleNet model here: [https://ngc.nvidia.com/catalog/models/nvidia:tlt\_peoplenet](https://ngc.nvidia.com/catalog/models/nvidia:tlt_peoplenet) The performance graph shows the Nano running inference with that model at zero frames per second, so maybe not a great choice for the Nano. ---
**Author:** ![dusty\_nv](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/dusty_nv/32/14043_2.png) [@dusty\_nv](https://forums.developer.nvidia.com/u/dusty_nv)\ **Post date:** [May 10, 2021, 8:27pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/9 "2021-05-10T20:27:06Z")
> [@loophole64](#): > > The performance graph shows the Nano running inference with that model at zero frames per second, so maybe not a great choice for the Nano. Thanks for pointing that out - you can find a table of the data here: [https://docs.nvidia.com/metropolis/TLT/tlt-user-guide/text/purpose-built\_models.html#deployment](https://docs.nvidia.com/metropolis/TLT/tlt-user-guide/text/purpose-built_models.html#deployment) PeopleNet is 14FPS for PeopleNet-ResNet18 and 11FPS for PeopleNet-ResNet34 on Nano. ---
**Author:** ![loophole64](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/loophole64/32/38308_2.png) [@loophole64](https://forums.developer.nvidia.com/u/loophole64)\ **Post date:** [May 10, 2021, 8:58pm UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/10 "2021-05-10T20:58:52Z")
> [@dusty\_nv](#): > > Thanks for pointing that out - you can find a table of the data here: > > [https://docs.nvidia.com/metropolis/TLT/tlt-user-guide/text/purpose-built\_models.html#deployment](https://docs.nvidia.com/metropolis/TLT/tlt-user-guide/text/purpose-built_models.html#deployment) > > PeopleNet is 14FPS for PeopleNet-ResNet18 and 11FPS for PeopleNet-ResNet34 on Nano. That sounds much better! Thanks Dusty. ---
**Author:** ![nadeemm](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/nadeemm/32/367020_2.png) [@nadeemm](https://forums.developer.nvidia.com/u/nadeemm)\ **Post date:** [October 15, 2021, 3:37am UTC](https://forums.developer.nvidia.com/t/jetson-nano-best-person-detector-model/173845/11 "2021-10-15T03:37:20Z")