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.
197 questions
0
votes
0
answers
5
views
How to deal with many unlabeled data in an object detection dataset
I've a large multi-class object detection image dataset. The goal is to use a Yolo(v11) model to be trained on the aforementioned dataset to solve the object detection task.
My intuition says that the ...
- 337
1
vote
0
answers
9
views
How does one measure the fall in performance of Object Detection algorithms with degraded training images? [closed]
I'm supposed to be working on a project that involves studying of the effect on Object Detection algorithms with different kinds of training images. We(team) are supposed to measure the difference in ...
- 11
0
votes
0
answers
16
views
How to judge if some prediction results are good in this particular case
I would like some advice on my particular case.
I have some script that analyze some Ground truth and some prediction results on object detection.
The results from the analysis is a csv file of this ...
- 257
0
votes
0
answers
56
views
Average precision vs Average recall in object detection
There are two popular metrics for object detection: Average precision and Average recall. Do you can explain with examples, what are the cases to use AP, and what are the cases to use AR?
I agree that ...
- 31
2
votes
1
answer
35
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 ...
1
vote
0
answers
45
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 ...
0
votes
0
answers
82
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 ...
- 1
0
votes
0
answers
11
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 ...
0
votes
0
answers
26
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....
0
votes
0
answers
99
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 ...
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 ...
- 11
1
vote
0
answers
56
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 ...
- 111
1
vote
0
answers
15
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!
- 11
1
vote
0
answers
25
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 ...
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 ...
- 37
2
votes
1
answer
462
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 (...
- 21
1
vote
0
answers
156
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 ...
- 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 ...
- 21
2
votes
0
answers
308
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 ...
- 183
4
votes
0
answers
1k
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}\ ...
- 41
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 ...
- 41
3
votes
0
answers
415
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 ...
- 31
0
votes
1
answer
435
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 ...
- 1
1
vote
0
answers
168
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 ...
- 11
1
vote
0
answers
17
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 ...
- 11
1
vote
0
answers
886
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 ...
- 11
1
vote
2
answers
787
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 ...
- 845
-1
votes
1
answer
857
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 ...
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?
1
vote
0
answers
84
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 ...
1
vote
0
answers
38
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 ...
- 51
1
vote
0
answers
110
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 ...
- 11
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 ...
- 23
1
vote
0
answers
90
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 ...
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 ...
1
vote
0
answers
87
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 ...
- 11
0
votes
0
answers
507
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 ...
- 1
1
vote
0
answers
62
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 ...
- 103
1
vote
1
answer
477
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 ...
- 13
1
vote
1
answer
340
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
- 11
2
votes
1
answer
870
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 ...
- 501
2
votes
0
answers
111
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 ...
- 1,116
3
votes
1
answer
422
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 ...
- 1,116
2
votes
0
answers
119
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 ...
- 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 ...
- 11
0
votes
1
answer
234
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, ...
- 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 ...
- 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 ...
- 111
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 ...
2
votes
1
answer
710
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 ...