Skip to main content

Questions tagged [object-detection]

Object detection deals with recognizing the presence of objects of a certain semantic class (e.g., “humans”, “buildings”, “cars”, etc.) in digital image and video data.

Filter by
Sorted by
Tagged with
2 votes
1 answer
27 views

Object detection for finite dataset

Consider the following scenario If I want to train a model to detect and count these squares: These squares will never be different. They will always look exactly the same, and be of exactly the ...
Fresh Prince Of Nigeria's user avatar
1 vote
0 answers
32 views

Is it reasonable to use background subtraction to identify some objects in sequential frame images to start labeling objects for YOLO training?

I know background subtraction is not a complete solution for object detection, but I’ve tried it for identifying potential new objects appearing in fixed background camera scenarios (millions of ...
NominalSystems's user avatar
0 votes
0 answers
33 views

How to Plot and Interpret the ROC Curve for Segmentation-based Object Detection Models?

I'm trying to plot the ROC Curve for a number of target/object detection models and compare their performance. The pre-trained models in question take an input image and they output a mask image where ...
Tungdil's user avatar
0 votes
0 answers
10 views

Wilcoxon's Signed-Rank Test in the context of 2 algorithms and 1 domain

I'm trying to understand whether my analysis for a problem is in the right direction. I have 2 algorithms (3d object detectors) that I've applied to the same dataset to obtain TP, FP and FN's for each ...
neoavalon's user avatar
0 votes
0 answers
19 views

Confidence interval for object counting

I'm working on object counting problem with deep learning object detection methods (specifically, yolo and faster rcnn). Is there any known method for uncertainty qunatification for object counting (i....
Andrew Lee's user avatar
0 votes
0 answers
71 views

Why regarding mean Average Precision for 50 and for 50 to 95? And can you average different evaluations?

I saw a lot of papers using mAP@50 and mAP0:5:95, but in terms of deciding how precise a trained object detection model is, I do not understand why often both values are discussed and shown. Wouldn't ...
Taka Incur's user avatar
0 votes
0 answers
11 views

Why detections count is not equal to unique truth count in YOLOv4 test result report?

I trained model using YOLOv4 on GPU, CUDNN and openCV (python) with AlexeyAB\Darknet with multi-label on windows environment. These labels are 25 classes (from 0 to 24). Then I test the model and I ...
N.white's user avatar
0 votes
0 answers
8 views

Object detection: better to train with imbalanced dataset or remove images to balance out [duplicate]

I am training an object detection model using the YOLOv5 architecture. I have the following classes and counts. ...
Peter's user avatar
  • 239
1 vote
0 answers
19 views

Probability of an image containing a specific object, by combining the results of multiple dependent tests

I am trying to assign to images the probability of them containing a metal building. The images can contain either metal buildings, non-metal buildings or no building. Let $B$ be the event that an ...
lovonho's user avatar
  • 11
1 vote
0 answers
46 views

Confidence intervals for Object Detection metrics

I would like to come back on this "When do we require to calculate the confidence Interval?" since recently a reviewer asked me to provide confidence intervals for metrics regarding my work ...
rok's user avatar
  • 111
1 vote
0 answers
12 views

Any image datasets with Inter-annotator agreement (IAA) values recorded? [closed]

Is anyone familiar with publicly accessible image datasets that include Inter-annotator agreement (IAA) scores? Ideally for object detection or classification tasks. Thank you!
SQL_Noob's user avatar
1 vote
0 answers
24 views

Object detection and geolocating the bounding boxes

I have acquired videos from a dashcam installed on a vehicle that went around different portions of an airport. My objective is to detect cracks in the pavements. After identifying them, I then need ...
Arman Malekloo's user avatar
2 votes
1 answer
1k views

Average Precision (AP) for object detection, huge confusion

I've been reading about how object detection models are evaluated. It seems that the metric most often used is AP. But I have stumbled upon 2 different approaches that I think mean completely ...
Tomé Silva's user avatar
2 votes
1 answer
427 views

For "fine-tuning", does the "domain adaptation" approach make sense?

I understand "domain adaptation" to be a type of "transfer-learning" technique. Domain Adaptation: By applying knowledge obtained from a domain with sufficient teacher labels (...
lee_0213's user avatar
1 vote
0 answers
147 views

YOLOv1 and YOLOv2 -- Pr(Object) and "responsible" boxes

In the YOLOv1 paper, Pr(Object) denotes the probability that there is an object in a certain grid cell. When it comes to the ground truth values, which values can Pr(Object) take? Only 0 and 1? What ...
keezar's user avatar
  • 35
1 vote
2 answers
2k views

How to count number of false positives and false negatives in object detection

I've got a rather simple question for you, however to which I can't find a proper answer. Let's have a simple setting which is visualized in the image below. We've got one object of class A in the ...
marveso's user avatar
  • 21
2 votes
0 answers
277 views

Guidelines for using NMS before calculating mAP for object detectors

I am having a hard time understanding how to use Non-Max Suppression (NMS) when trying to evaluate an object detection model, especially when paired with trying to calculate metrics like the mean ...
Will.Evo's user avatar
  • 183
4 votes
0 answers
992 views

The alpha parameter of focal loss

I want to use a weighted focal loss for my imbalanced object detection problem. \begin{equation} L = - \alpha(1-\hat{p})^\gamma log(\hat{p}) , \ \hat{p} = \begin{cases} p, & \text{if}\ ...
The_Tortoise's user avatar
4 votes
1 answer
1k views

Training with extremely imbalanced Dataset

I have a object detection problem which has extremely imbalanced dataset. Lets say there is only one class to detect, say banana or not banana. This detection network will be used in a real case where ...
Uce's user avatar
  • 41
3 votes
0 answers
408 views

I am trying to do object detection using Yolov5 and modify the model. How can I do that?

I want to modify Yolov5 and see how it performs. Here's the yaml file https://github.com/ultralytics/yolov5/blob/master/models/yolov5s.yaml How do the make the changes in the yaml file if I want to ...
Lucie's user avatar
  • 31
0 votes
1 answer
387 views

Can bounding boxes be used in UNets?

I recently read a paper where the researchers used a UNet algorithm to localize/detect cyclones using a bounding box. However, my interpretation of a UNet is that it performs semantic segmentation and ...
lica's user avatar
  • 1
1 vote
0 answers
157 views

Yolov3 targets build, Do all 3 scales need to have an anchor box designed for each object in the image or just the scale with better IoU? [closed]

I'm implementing Yolo v3 in Pytorch, Imagine we have an image with a large object in it, whem building targets, I'll have to assign the anchor box with highest IoU with this object for each scale, (So ...
Vinc's user avatar
  • 11
1 vote
0 answers
16 views

"Biasing" a generalized trained object detector towards specific examples during inference

I have trained an object detection deep learning model on many different types of cars (shape, color, car model variations etc.). I'm just using a single class "Car" for all the different ...
Rahul's user avatar
  • 11
1 vote
0 answers
778 views

YOLO v2 loss function

I'm trying to understand (and implement) the YOLOv2 loss function, which is not given explicitly in the original paper. There are several posts on this topic, but quite a few seem to confuse the ...
eike's user avatar
  • 11
1 vote
2 answers
694 views

Why do we sort confidence scores when calculating MAP in Object Detection?

This question is specific to object detection's metric Mean Average Precision. I sieved through multiple articles like the one by Jonathon and here by Kaggler Tito. One part I cannot grasp immediately ...
nan's user avatar
  • 845
0 votes
1 answer
823 views

What should I do with Images that has no objects?

I have a dataset that contains images that has cancerous nodules. I want to use object detection models to detect these nodules from the image by using an object detection model. Now my dataset has ...
Tareq Mahmud's user avatar
1 vote
1 answer
1k views

Is Intersection over Union in object detection differentiable?

Is IoU in object detection differentiable or can be back propagated? Is it used in the training process or just in the inference?
Chandler Timm's user avatar
1 vote
0 answers
83 views

Annotations and Bounding Box Definition in Object Detection

I'm having confusion about the concepts of object detection. If the dataset has been annotated to have ground truth in (x, y) of the four vertices, then why are the papers (using the same dataset) are ...
Chandler Timm's user avatar
1 vote
0 answers
31 views

How are the region proposals processed in Fast R-CNN during training and testing?

How are the region proposals processed in Fast R-CNN during training and testing? From my understanding, during training we sample some small number (e.g. 64) of region proposals out of the ~2000 ...
orbit's user avatar
  • 51
1 vote
0 answers
94 views

Conditional class probabilities in YOLOv1

In YOLO v1, authors defines conditional class probabilities (P(Class_i/Object)) I don't understand that authors can define the head output to conditional class probabilities. When the model finishes ...
Jaehong choi's user avatar
1 vote
1 answer
3k views

When is Non-Max Suppression used in Object Detection

Is non-max suppression for bounding boxes obtained from a Region Proposal Network performed during training? From what I gather, NMS is not differentiable-- in which case, it can't be performed during ...
skinnybb's user avatar
1 vote
0 answers
88 views

how to check if unknown object partly covers the object i want to detect

I want to detect an object from an image,I use yolo object detection. I detect car from the image. I consider the case when another unknown object partly covers it, how can my program know that there ...
anahit hambardzumyan's user avatar
0 votes
1 answer
1k views

Object detector evaluation (multi-class)

I have a (long) question regarding the evaluation of object detectors. I have two separate networks, the first one detects relevant objects (-> bounding boxes) and the second one classifies the ...
python1235432's user avatar
1 vote
0 answers
85 views

Fine-tuning a pre-trained Tensorflow model

I have two questions regarding fine-tuning a pre-trained Tensorflow model on a mix of high-quality and low-quality images.In other words, ' Does the large contrast in image quality confuses the model ...
learner 's user avatar
0 votes
0 answers
492 views

Template matching for detecting multiple objects in image (+500 different categories)

I want to detect items from the inventory of a game. Detection is in real time, so I need decent performance (> 5 FPS). There are over 500 different items, but the item icons will not change so a ...
igonro's user avatar
  • 1
1 vote
0 answers
58 views

Summary of all the results of all participants after 2015 in ImageNet (ILSVRC) challenge. Does such resource exist?

I have been reading about the ImageNet Large Scale Visual Recognition Challenge (ILSVRC). And while I can find the paper by Russakovsky et al. from 2014(updated in 2015), which contains all the ...
vcucu's user avatar
  • 73
1 vote
1 answer
450 views

Performing object detection without localization

I'm looking to build a computer vision model that can detect the presence or absence of several classes of object within an image, but the localization of these objects is not required. The end ...
Joe Kerrigan's user avatar
1 vote
1 answer
328 views

What is the difference between these error metrics in object detection

I am wondering what is the difference in these error metrics (and what they are) in object detection: mAP_0.5:0.95, mAP_0.5, recall, precision
Peter's user avatar
  • 11
2 votes
1 answer
839 views

Does it help to provide training samples with no target objects in object detection?

I am training a Yolo for object detection. I was wondering if it will help to provide images with no instances of the target object present. If so, what would the VOC label of such an instance look ...
maininformer's user avatar
2 votes
0 answers
107 views

what does it mean that the effective receptive field to be large for Region Proposal Network in Faster R-CNN?

I am trying to understand the Region Proposal Network portion (RPN) of the faster R-CNN paper. Under section 3.1, the authors mention the justification for using a 3x3 sliding network over the ...
calveeen's user avatar
  • 1,116
3 votes
1 answer
400 views

Mean Average Precision (MAP) object detection metric

I am reading up on object detection metrics from this github repository. I have a slight confusion regarding the precision x recall curve mentioned under the Metrics heading. It says that The ...
calveeen's user avatar
  • 1,116
2 votes
0 answers
109 views

CNN Color Invariance

To detect a specific object that can have any color, if we train our model with images containing that object with just a few colors (e.g. 2 or 3 colors), will the performance of our model's ...
TheDude's user avatar
  • 21
1 vote
0 answers
134 views

YoloV3: Training set, true label boxes vs anchor boxes

I understand the anchor boxes as a set of boxes that based on a comprehensive exploration of the data ( all the true bonding boxes) best describe the variable sizes of all true boxes in the training ...
user317766's user avatar
0 votes
1 answer
222 views

Is there any statistically meaningful definition for object confidence in object detection?

Most modern object detection algorithms rely on neural networks and output a bounding box and confidence for each object (or more accurately, a confidence for each possible object class considered, ...
DerekG's user avatar
  • 246
3 votes
1 answer
2k views

What are "Grids" and Detection at different scales" in YOLOV3?

I've recently started working with Yolov3 and the more I go in depth, the more confused I get. In the simplest terms what I think about YOLOV3 ...
Deshwal's user avatar
  • 244
1 vote
0 answers
16 views

Can I conclude from object detection metrics on temporal behavior of a detection system?

Assuming I know typical object detection metrics like precision, recall etc. for a certain class, is there anyway to conclude on the probability that an object of that class which is present in ...
P. Leibner's user avatar
0 votes
1 answer
1k views

What is the average precision in the case of no positives for a given category in the context of object detection

In attempting to calculate the average precision of an object detection model, I am wondering about an edge case. Suppose at evaluation time that for a given category, that no detections of that ...
IntegrateThis's user avatar
2 votes
1 answer
690 views

When computing mAP for an object detection model, how many detections should one consider?

I am trying to write some code to evaluate the MS COCO style mAP (mean average precision, average computed at the category level) at different IOU levels in the context of object detection with a ...
IntegrateThis's user avatar
1 vote
2 answers
1k views

Two True Positive for one ground truth in object detection

I am wondering is it possible to have two true positive predictions for one bounding box ground truth only. Following this section from Stanford. They define truth positive like this: We start with ...
CuCaRot's user avatar
  • 113
2 votes
0 answers
245 views

Enemy detection in 3D video games without labeled dataset

I'm working on a tool for single-class object detection in video games, which should be able to detect enemies in 3D environments. The main problem is, that I cannot use any labeled dataset. Instead I ...
Bob's user avatar
  • 21