Use "detectron" tag for questions related to FAIR's detectron or detectron2 framework for object detection and segmentation.
Questions tagged [detectron]
217 questions
0
votes
0 answers
How to train a detectron2 model without bounding box in the training set but have the segmentation labels
I have a training dataset which has many images but do not have bounding box annotations.
I have annotations for the images in validation and test dataset.
Goal is to train a detectron2 model using validation (with bbox) and training dataset(without…
AMRUTA BADGUJAR
- 43
- 1
- 4
0
votes
1 answer
got "Could not export Python function call '_ScaleGradient'" when export detectron2 model to torchscript
After trained a module by Detectron2, I tried to export the model to TorchScript,
Then I got the following errors:
Could not export Python function call '_ScaleGradient'. Remove calls to Python functions > before export. Did you forget to add…
Ben
- 9
- 4
0
votes
1 answer
Image size in DefaultPredictor of Detectron2
For object detection, I'm using detectron2.
I want to fix the input image size so I made my customized dataloader:
def build_train_loader(cls, cfg):
dataloader = build_detection_train_loader(cfg,
mapper=DatasetMapper(cfg, is_train=True,…
Kosar KAZEMI
- 3
- 1
- 2
0
votes
1 answer
Torch is installed but I'm unable to import it in a computer vision python project in Jupyter notebook
I'm working on a computer vision project using detectron2. I'm having trouble installing torch or importing it into my jupyter notebook. I'm working on a mac that is running MACOS Catalina, Python3 version in 3.8.2 and I'm using Anaconda for my…
Muhammad Sameer
- 27
- 7
0
votes
1 answer
Give randomCrop augmentation and loss become explode
I'm trying to test the different kind of augmentation,
but when I gave option with RandomCrop it gives loss value NaN or infinity.
Here is my random augmentation optims
def mapper2(dataset_dict):
dataset_dict = copy.deepcopy(dataset_dict) # it…
user14556788
0
votes
1 answer
AttributeError: Cannot find field 'pred_masks' in the given Instances
I trained a model with RetinaNet Detectron2 on colab. The training is successful but i am unable to test it on images.
Code:
from detectron2.utils.visualizer import ColorMode
import glob
for imageName in glob.glob('/content/drive/My…
Kranthi Kumar
- 1
- 5
0
votes
1 answer
Empty output file, Detectron2
I want to create body pose estimator with Detectron2. I have been trying to run following code locally to read a video file, make the prediction frame by frame and record a video with the processed frames. But it doesn`t work properly. It creates…
Teodred
- 1
- 1
0
votes
1 answer
In detectron2, how could I put the text on image?
I am learning how to use detectron2 well. And I could do predicting the bounding box. But, I also want to put the bounding box coordinate on the image. For this, I use cv2.putext library. but it did not work.
Could you please make the below code can…
Seunghyeon
- 103
- 1
- 2
- 10
0
votes
1 answer
How to use the Detectron2 .pth model for prediction from storage..?
I have trained the Detectron2 model on the google colab server free server.
from detectron2.engine import DefaultTrainer
from detectron2.config import get_cfg
import os
cfg = get_cfg()
cfg.merge_from_file(model_zoo.get_config_file("COCO-…
JB.py
- 59
- 2
- 8
0
votes
1 answer
How can I plot my trained model result on video using Detectron2?
I am new on using Detectron2. I want to load the video from local drive. And then, do detection using my trained model using Detectron2's VideoVisualizer.
I tried to find a tutorial about this. But it does not exist. Could you please what do I…
Seunghyeon
- 103
- 1
- 2
- 10
0
votes
1 answer
Detectron2 video object detection
I run obeject detection code, using detectron2.
I trained the net and it works fine on images.
Now I would like to test it on video, but when I run the video it doesn't detect any of my labels, more then that, it detect untrained object, means my…
alik
- 21
- 1
- 5
0
votes
1 answer
How do I get X server to display an image in detectron2?
I am following detectron2 getting started tutorial to detect objects in an image using one of their machine learning models. I am doing this using an AWS machine learning AMI and jupyter notebook.
The model successfully detects an object in the…
0
votes
1 answer
Storing given percentage of shape contour coordinates
I am building a project on Detectron2 with some mushrooms as a topic. The prediction works OK-well and I'm now trying to generate COCO-like annotations of the images with the predicted region (all XY coordinates of the region). For this, I need to…
Stefano
- 43
- 4
0
votes
1 answer
Multi-object tracking with dlib
I'm trying to implement the following code from the pyimagesearch web for multi object tracking with dlib. I tried to customize it to run it with detectron2 instead of the Mobilenet + SSD Caffe model. Here is the code
#!/usr/bin/python
# -*- coding:…
Laura_M
- 43
- 1
- 3
-1
votes
2 answers
Error with Detectron2 : RuntimeError: Input type (torch.cuda.DoubleTensor) and weight type (torch.cuda.FloatTensor) should be the same
I'm getting this rather annoying error and I have no idea how to correct : RuntimeError: Input type (torch.cuda.DoubleTensor) and weight type (torch.cuda.FloatTensor) should be the same.
I'm running the code for image detection locally on a Jupyter…
NicolasRx
- 29
- 5