It’s my first time using, RealSense Depth Camera D435i, and so I’m excited about the potential of this camera! For example, it would make detection tasks much easier because you can literally distinguish between objects by the depth measured from…
Auxiliary Classifier is a classifier head which is not connected to the end of the network. This is helpful to improve convergence during training, because by having an output near the lower-level side of the network, the gradients can be…
There are several ways to evaluate your semantic segmentation model, and one way is to calculate the IOU(Intersection Over Union). This is basically calculating how much area is intersecting between the region proposed by the model and the actual region.…
I was finally able to fine-tune a model and get decent results. I’ve used a new dataset from kaggle. This has more and simple images of water surfaces so that helped a lot. Results The model was able to learn…
I trained L-RASPP again to detect cracks with 110 images.The results are the above. Comparing to Deeplabv3(Image Below),trained with the same dataset, you can see there are a lot of noise in the results from the L-RASPP model. > Reference:…
I retrained a L-RASPP(Light-Reduced Astrous Spatial Pyramid Pooling) model to detect water level using Pytorch. I used the WSOC dataset containing 496 images for training and 278 images for testing. The ROC score was close to 1 by epoch 25.…
Transposed Convolutional Layer is doing the following: Stretch Image by inserting rows and columns with zeros. Perform regular convolution You can think of it as a fractional stride, for example, 1 This is often used when doing semantic segmentation, where…
Here is 1 way to Generate an engine file with Unet Pre-requisites – Deepstream 5.0.1 – TensorRT 7.1.3 – Jetson XAVIER NX – CUDA 10.2 Load Model import torch unet = torch.hub.load(‘milesial/Pytorch-UNet’, ‘unet_carvana’, pretrained=True) unet.eval() Export Model path_to_onnx =”” dummy…
Before you deploy your model, you might want to run a test inference with the exported file, just in case to confirm your model was exported successfully. Here is how you can do it for Deeplab. NOTE: I am running…
Yeah, this wasn’t supposed to happen. I trained an Instance segmentation model using NVIDIA’s Transfer Learning Toolkit to detect cars and deployed it on a Jetson. It seems I messed up with either the settings for the threshold or with…