Questions tagged [object-detection]
Object detection is a computer-vision and image-processing technique for locating instances of objects in images or videos. Common applications include face detection and object tracking. Object detection algorithms typically leverage machine learning or deep learning to produce meaningful results.
352
questions
0
votes
0
answers
6
views
Installation of Tensor flow Object Detection API on CPU
I am trying to install the Tensorflow Object Detection API on local CPU . I am using python 3.11.4 and Downloaded Protobuf version 3.20.3 and added the protoc.exe file to the scripts of the virtual ...
0
votes
0
answers
13
views
How to Adjust parameters for training Tensor flow EfficientDet d0 model on a very small dataset
I am trying to extract details like invoice number, invoice date, total amount from multi layout pdf files. I have about 260 images in total for dataset after data argumentation. I am training ...
-1
votes
0
answers
20
views
Improving ml model for invoice extraction
I have to build a ml model for extracting invoice details like customer's name, invoice date , invoice number and others from different layouts of pdf files .I have small dataset of 252 images where ...
0
votes
1
answer
12
views
How could I further improve my detection model?
I am currently tasked with the creation of a model which is able to detect burrows of small rodents from high resolution aerial images.
My labeled images are look like this:
My results now are the ...
0
votes
1
answer
10
views
Object detection but only classification and not drawing bbox
I need to train a model that can tell what are the objects in the image but don't need to draw the bbox. Does my dataset still need bbox information?
Example of the inference function ->
inference(&...
0
votes
0
answers
33
views
Understanding re identification in BoT SORT in multi object tracking
I was trying to use BoT-SORT with reid on a simple video in which single person is walking on the road, first gets occluded by small tree and then by a billboard. Also this is a drone footage though ...
- 199
0
votes
0
answers
14
views
Making ReID work with BoTSORT object tracking
Am trying BoT-SORT repo.
I tried following commands to try out ReID on my custom 15 seconds video with three people walking, getting occluded behind two trees and emerge back from behind tree.
Single ...
- 105
0
votes
0
answers
17
views
How to train and test a weed detection algorithms using drone data
I have come across a weed database with labels and annotations for cotton weeds that was captured by a drone. Now I want to train and test the dataset which can be found here (Weed database). I do not ...
- 1
0
votes
0
answers
8
views
Understanding MOTChallenge dataset format
I was looking into MOT17 datasets. And I have some stupid questions regarding dataset:
Q1. Why there is no ground truth files in test datasets? For example ...
- 105
0
votes
0
answers
19
views
Does it make sense to have object detection model followed by a classification model
So i was working with the SKU110k dataset and i was required to identify the different items in the shelf as well but the SKU110k dataset only annotated shelf items but did not identify them. So i ...
0
votes
1
answer
21
views
What is the "fast version" of ZFNet referenced in SPPNet and Faster R-CNN papers?
I'm reading old papers:
SPPNet: Link
Faster R-CNN: Link
In both cases, the authors refer to a "fast version of Zeiler and Fergus (ZF) Net"; specifically:
In SPPNet:
ZF-5: this ...
1
vote
1
answer
21
views
Tracking people problem
I am trying to implement a system for counting people in a building using Python and OpenCV, but I have one problem. The system for tracking people doesn't work very well; it can't track a person all ...
- 11
1
vote
0
answers
53
views
How to implement hard negative mining for PyTorch RetinaNet_v2 object detection model?
I am currently using the PyTorch object detection model retinanet_resnet50_fpn_v2. I want to implement hard negative mining, such that I take the top X images with ...
- 11
0
votes
0
answers
10
views
ways to calculate the accuracy of a object detection model
In case of classification, we will calculate the accuracy by performance of model on test dataset.
...
2
votes
1
answer
49
views
How to chose the right activation function for CNN output depending on the output value ranges?
I'm working on training a CNN model that takes an eye image as input and outputs the 5 coordinates of the ellipse representing the pupil ...
- 23
0
votes
0
answers
23
views
Convolutional Filters that detect number of ones in binary matrix
I am studying a deep learning course and one of the questions in the course is like a puzzle game:
Part (a)
Given a 4 by 4 binary matrix (consisting only zeros and ones), design a CNN that can detect ...
- 1
0
votes
0
answers
12
views
Can a multi-task model work on conditions for each layer?
I am currently working on a multi-task model that needs to handle car detection, damage segmentation and license plate ocr.
my idea was to only run the OCR layer when there is a license plate detected ...
0
votes
0
answers
56
views
One shot object detection
I want to detect objects from a video.
I can only have one example image and based on this, I want to detect whether this image is present in the video or not.
What could be the best-suited approach?
0
votes
1
answer
28
views
How to interpret annotation data?
I am new to datasets. I have got an annotations train.json for MR image data like this -
I want to train a Yolo-V8 model using this MR data images extracted from dicom raw data and annotations for ...
- 168
1
vote
0
answers
21
views
Improving Detection Model - Adding image clarification
I trained an object detection model with 5K images, it works most of the time, but I am facing an issue, for few times, the object is not getting detected.
So, I planned to retrain the model, for that ...
- 21
0
votes
0
answers
8
views
CNN to estimate a density map of an image, used to predict the object count
I'm trying to count the number of objects (larvae in this case), from a video. The constraint is, I cannot use any ML model, nor can I train my dataset as there are no annotations available.
This ...
- 101
0
votes
0
answers
9
views
Why do object detection model adversarial masks look different from those of image classifiers?
I was messing around to observe the behavior for adversarial attacks on image classifiers, and decided to try it with an object detector as well. I realize that inference time attacks are more complex ...
0
votes
0
answers
3
views
How to use anchor box priors in yolov2?
I saw the paper in yolov2 where they said they picked some anchor boxes to start with. But the output also predicts a the position, width and height of the box. So there's no way to make the neural ...
1
vote
2
answers
35
views
Image segmentations vs image detection
If I need to detect on an image some objects and we are only interested in counting them, between image segmentation and object detection which one would you think would yield best results in terms of ...
- 23
0
votes
0
answers
25
views
Tile color, shape, and type detection altogether
I'm trying to apply some detections/classification on the set of tiles we have. Specifically, I need to detect color (15 classes), pattern (25 classes- on the surface of tiles, there can be certain ...
- 217
0
votes
0
answers
35
views
How to treat "ignore" boxes during object detector evaluation?
I want to evaluate the performance of multiple object detectors for comparison, each of them has been trained on this data. This dataset has images labeled, where each image can have multiple labels. ...
- 121
1
vote
2
answers
307
views
Why not using segmentation architectures for object detection?
Current object detection architectures like Faster-RCNN and YOLO seem to be overcomplicated in comparsion with segmentation architectures like Unet. So, why can't we just draw some rectangles around ...
- 13
0
votes
1
answer
81
views
Optimizing YOLO for Diverse Takeaway Item Detection: Single vs Type-Based Classes
I'm using YOLO to detect various 'takeaway' items, currently all marked as class 0, which vary very widely(like groceries) in shape and type. Considering the introduction of new items on which the ...
0
votes
1
answer
934
views
Car Make and Model detection
I am trying to develop a deep learning model that given an image of a car, it detects a car's make and model among 50 different brands, each with say another 50 models. What approach is probably the ...
- 217
0
votes
0
answers
26
views
Tensorflow outputs nan for basic object detection/classification
I am receiving nan as my accuracy and loss outputs after each epoch for basic object detection in tensorflow. Also, my results (classification and bounding box ...
- 1
0
votes
0
answers
27
views
IoU when labels are different
The IoU focuses only on the bounding boxes.
My predictions for one Image are (Yolo) e.g.
label
x
y
w
h
0
0.1
0.2
0.3
0.3
1
0.9
0.9
0.05
0.05
And my ground truth is:
label
x
y
w
h
1
0.1
0.2
0.3
0....
- 101
1
vote
0
answers
329
views
Accuracy difference between 1-channel grayscale and 3-channel grayscale detection model
I have found no similar questions to this online, or answers for that matter.
I am using cameras that output a grayscale image, which I feed into a Yolov8 object detection model (Specifically yolov8m-...
0
votes
0
answers
12
views
Identify multiple people in images and count the frequency
I have multiple images posted by a user and I am interested to know how many times each unique person appears in the user's photos. For instance, the user might post photos of himself/herself with ...
- 1
0
votes
0
answers
107
views
Torchvision Faster-RCNN, modified loss function
I'm trying to solve a problem of table detection in spreadsheets in Excels. I've came across this paper, which suggests to use modified version of Faster RCNN to do object detection on the ...
- 101
0
votes
1
answer
27
views
Detector vs Classification - Detecting a netted bag for fruit/vegetables in image
The application is detecting the presence of a netted bag in an image.
The image can contain fruit and vegetables, either with or without a netted bag around them, or below them (no constraints about ...
1
vote
0
answers
19
views
Can CNNs complete lines and contours?
Are there deep convolutional networks capable of recognizing two overlapping triangles in this image - or is this beyond the capabilities of CNNs?
And are there CNNs that can recogize two boxes ...
0
votes
1
answer
55
views
MaskRCNN.train gives 'list index out of range'
I have been trying to use MaskRCNN with a Resnet backbone on the DeepFashion2 Dataset for instance segmentation. The custom configurations are as follows:
...
- 23
0
votes
1
answer
91
views
Use computer vision to detect door blockage
I want to detect door blockage on a camera. Basically if the exit door is blocked by an object, it detects it as an anomaly. How can we do it? Is it possible to do it using OpenCV?
Remember, it doesn’...
- 217
1
vote
0
answers
79
views
Annotating and Structuring a Dataset for Duplicate Detection
I'm currently working on a project that requires the detection of duplicate bands in Western blot images. The task involves two types of duplicates:
...
- 111
1
vote
0
answers
186
views
Where exactly in YOLO's architecture is the input image divided into a grid?
I am currently studying the YOLO algorithm for a project. What I'm not quite sure about is where exactly the input image is divided in an SxS grid. After my research on the paper, videos and websites ...
- 11
0
votes
1
answer
35
views
Best practice labeling grouped anomalies for object detection
I would like to train object detection model (e.g. YOLO) for images that contain anomalies. The anomalies are essentially the holes in a surface of different sizes. How do I label correctly such ...
- 3
0
votes
1
answer
1k
views
How does the background class work in object detection?
I am using YOLOv5 for object detection.
I understand that any labelled classes that are not predicted, that is, false negatives (FN) shows up as background. But how are the false positive (FP) being ...
1
vote
0
answers
24
views
How to associate each object in the prediction to its annotated version in YOLOv5?
Imagine we have one image from the COCO validation dataset. This image has 7 annotated objects: knife, carrot, person, table, chair, apple, and orange. When I feed it to YOLOv5, the predictions are ...
- 123
1
vote
1
answer
49
views
Can we create tensorflow or tflite model for object detection without uploading our dataset images to cloud?
I need to create a custom Tensor flow lite model for object detection to integrate in an android app. But I have a constraint that the dataset images to be used is confidential and cant be uploaded ...
0
votes
1
answer
52
views
research papers on mean average precision
I've been coding YOLOv1 from scratch, are there any good papers which explain and/or give code (or pseudocode) for mean average precision? I searched but couldn't find good ones
1
vote
1
answer
755
views
Object Detection: Setting threshold values as trainable parameters?
I am building my first object detection model (Mobilenet SSD, to detect animals in images) and happy with the current test results.
When I tested it using images without bounding boxes, I noticed some ...
- 11
5
votes
3
answers
392
views
Spatial Join Pandas Dataframes of Bounding Boxes (cross match)
Problem Statement
Imagine there are two almost identical images with annotations (bounding boxes for certain objects), where one is so-called golden image (template) containing all must-have objects (...
- 4,299
0
votes
1
answer
2k
views
Yolov8 - box_loss and dfl_loss stays at 0. cls_loss converges. Model not giving me bounding box predictions
I'm having trouble using Yolov8 to work properly. I have my own custom dataset and an online dataset that I am using. Yolov8 trains on these datasets. However, the only metric that converges is the ...
- 1
1
vote
0
answers
146
views
Memory usage of a DNN model during inferencing
I am running a DNN model (YOLOv7) on GPU to detect objects in a video stream. The memory usage on my task manager shows that I am using up to 2GB of my RAM! The GPU Ram usage is almost 1 GB.
I am ...
- 11
0
votes
2
answers
72
views
how to lower false positive ratio on object detection using negative examples
There is a similar question here but the answer is not so clear;
Basically, I have a model that detects only and only "matchbox". However, it has a high false positive ratio specially ...
- 343