Adding chapter for object detection. #202
Conversation
| @@ -0,0 +1,63 @@ | |||
| # Object Detection: | |||
|
|
|||
| In this guide, we'll explore the fascinating world of object detection—a vital component in modern computer vision systems. We'll demystify essential concepts, discuss popular methods, examine applications, and touch upon evaluation metrics. By the end, you'll have a solid foundation and be ready to venture further into advanced topics. | |||
There was a problem hiding this comment.
I think that we shouldn't use "guide"
There was a problem hiding this comment.
Cool, I will change that.
|
|
||
| Understanding object detection lays the groundwork for mastering advanced computer vision techniques, enabling the construction of powerful and accurate solutions addressing rigorous needs. Some future research areas include developing lightweight object detection models which are fast and easily deployable. Exploration in the field of object detection in 3D space, e.g., for augmented reality applications, is another avenue to explore. | ||
|
|
||
| ## 8. References and Additional Resources |
There was a problem hiding this comment.
adding common datasets?
There was a problem hiding this comment.
In the references? Datasets like VOC, MS-COCO?
There was a problem hiding this comment.
@johko What do you think about it?
|
Hi! |
|
|
||
| || | ||
| |:--:| | ||
| | *Image displaying the bounding boxes around multiple objects in the frame along with the confidence score of their classification* | |
There was a problem hiding this comment.
you can write this in alt text instead of making this in a table format
There was a problem hiding this comment.
@merveenoyan , I wanted it to be like a caption to the image, alt text won't be visible I believe. What do you think?
There was a problem hiding this comment.
I think when people scroll on it they'll see so it should be ok
|
@merveenoyan , I have added the suggested changes, except for a few. I have added comments for that. Please check and let me know if it's fine. |
|
@miniMaddy I answered them, do you have any other questions |
|
Thanks a lot, I will push the changes by tonight then. |
|
All the comments are resolved. Please check and merge. Thank You |
|
|
||
| ### 1.2 Classification vs Localization | ||
|
|
||
| Classification distinguishes objects based on their unique attributes, while localization determines an object's location within an image. Object detection combines both approaches, encapsulating identified entities with bounding boxes and assigning corresponding class labels. Imagine recognizing different fruit types and pinpointing their exact locations in a single image. That's object detection at play! |
There was a problem hiding this comment.
Not all object detection involves using bounding boxes. It not unusual (at least in the biomedical field) to predict points within-in regions and centroids.
|
|
||
| ### 5.1 Intersection over Union (IoU) | ||
|
|
||
| Intersection over Union (IoU) measures the overlap between predicted and actual bounding boxes as a percentage ranging from 0% to 100%. Higher IoU percentages indicate better alignments, i.e., improved accuracy. Useful when assessing tracker performance under changing conditions, e.g., following wild animals during migration. |
There was a problem hiding this comment.
| Intersection over Union (IoU) measures the overlap between predicted and actual bounding boxes as a percentage ranging from 0% to 100%. Higher IoU percentages indicate better alignments, i.e., improved accuracy. Useful when assessing tracker performance under changing conditions, e.g., following wild animals during migration. | |
| Intersection over Union (IoU) measures the overlap between predicted and reference labels as a percentage ranging from 0% to 100%. Higher IoU percentages indicate better alignments, i.e., improved accuracy. Useful when assessing tracker performance under changing conditions, e.g., following wild animals during migration. |
|
|
||
| ### 5.2 Mean Average Precision (mAP) | ||
|
|
||
| Mean Average Precision (mAP) estimates object detection efficiency using both precision (correct prediction ratio) and recall (true positive identification ability). Calculated across varying IoU thresholds, mAP functions as a holistic assessment tool for object detection algorithms. Helpful when measuring sentiment analysis models' effectiveness, e.g., processing extensive customer reviews. |
There was a problem hiding this comment.
could you change this to a computer vision example?
| In this guide, we'll explore the fascinating world of object detection—a vital component in modern computer vision systems. We'll demystify essential concepts, discuss popular methods, examine applications, and touch upon evaluation metrics. By the end, you'll have a solid foundation and be ready to venture further into advanced topics. | ||
|
|
There was a problem hiding this comment.
| In this guide, we'll explore the fascinating world of object detection—a vital component in modern computer vision systems. We'll demystify essential concepts, discuss popular methods, examine applications, and touch upon evaluation metrics. By the end, you'll have a solid foundation and be ready to venture further into advanced topics. | |
| In this guide, we will explore the fascinating world of object detection — a vital component in modern computer vision systems. We will breakdown essential concepts, discuss popular methods, examine applications, and touch upon evaluation metrics. By the end, you will have a solid foundation and be ready to venture further into advanced topics. | |
|
|
||
| ### Classification vs Localization | ||
|
|
||
| Classification distinguishes objects based on unique attributes, while localization determines an object's location within an image. Object detection combines both approaches, encapsulating identified entities with bounding boxes and assigning corresponding class labels. Imagine recognizing different fruit types and pinpointing their exact locations in a single image. That's object detection at play! |
There was a problem hiding this comment.
| Classification distinguishes objects based on unique attributes, while localization determines an object's location within an image. Object detection combines both approaches, encapsulating identified entities with bounding boxes and assigning corresponding class labels. Imagine recognizing different fruit types and pinpointing their exact locations in a single image. That's object detection at play! | |
| Classification distinguishes objects based on unique attributes, while localization determines an object's location within an image. Object detection combines both approaches, locating entities and assigning corresponding class labels. Imagine recognizing different fruit types and pinpointing their exact locations in a single image. That's object detection at play! |
I am really sorry I forgot to commit my comments for two reviews last week 🤦. I had some minor suggestions just to keep things slightly more general. Otherwise, looks good to me :) |
|
Thanks @bellabf , I made the suggested changes. Thanks for correcting me on the detection boxes. |
|
merging 🚀 |
This is the chapter for object detection. I have just added small changes to the work already done by @abdullah-alnahas. Please suggest any changes if necessary.