Questions tagged [object-detection]

For questions related to object detection (where objects can be e.g. humans, dogs, houses, etc.), whose meaning or definition can vary depending on the context. OD can refer to the task of locating (i.e. finding the coordinates) an object in an image (so, in this case, it would be a synonym for object localization) or the task of locating the object and classifying it (i.e. object localization + object classification).

Filter by
Sorted by
Tagged with
11
votes
3answers
10k views

Is it difficult to learn the rotated bounding box for a (rotated) object?

I have checked out many methods and papers, like YOLO, SSD, etc., with good results in detecting a rectangular box around an object, However, I could not find any paper that shows a method that learns ...
9
votes
1answer
254 views

Extending FaceNet’s triplet loss to object recognition

FaceNet uses a novel loss metric (triplet loss) to train a model to output embeddings (128-D from the paper), such that any two faces of the same identity will have a small Euclidean distance, and ...
6
votes
1answer
168 views

Formal definition of the Object Detection problem

For many problems in computer science, there is a formal, mathematical problem defition. Something like: Given ..., the problem is to ... How can the Object Detection problem (i.e. detecting objects ...
5
votes
1answer
117 views

How does the region proposal method work in Fast R-CNN?

I read so many articles and the Fast R-CNN paper, but I'm still confused about how the region proposal method works in Fast R-CNN. As you can see in the image below, they say they used a proposal ...
5
votes
1answer
70 views

Do models train better if the labelling information is more specific (or dense)?

I'm working on a project where there is a limited dataset of videos (about 200). We want to train a model that can detect a single class in the videos. That class can be of multiple different types of ...
5
votes
2answers
2k views

What's the role of bounding boxes in object detection?

I'm quite new to the field of computer vision and was wondering what are the purposes of having the boundary boxes in object detection. Obviously, it shows where the detected object is, and using a ...
5
votes
1answer
142 views

Should I train different models for detecting subsets of objects?

Suppose we have $1000$ products that we want to detect. For each of these products, we have $500$ training images/annotations. Thus we have $500,000$ training images/associated annotations. If we want ...
5
votes
0answers
113 views

Are there any easy ways to create annotated training images for object detection?

For the purposes of object detection, are there any easy ways to create annotated training images? For example, if we have $10,000$ images and want to draw bounding boxes on 2 objects for each image, ...
4
votes
1answer
118 views

Why object detection algorithms are poor in optical character recognition?

OCR is still a very hard problem. We don't have universal powerful solutions. We use the CTC loss function An Intuitive Explanation of Connectionist Temporal Classification | Towards Data Science ...
4
votes
1answer
320 views

What loss function should one use for object detection, knowing that the input image contains exactly one target object?

What loss function should one use, knowing that the input image contains exactly one target object? I am currently using MSE to predict the center of ROI coordinates and its width and height. All ...
4
votes
0answers
957 views

What are the differences between Yolo v1 and CenterNet?

I recently read a new paper (late 2019) about a one-shot object detector called CenterNet. Apart from this, I'm using Yolo (V3) one-shot detector, and what surprised me is the close similarity between ...
3
votes
1answer
77 views

How to treat (label and process) edge case inputs in machine learning?

In every computer vision project, I struggle with labeling guidelines for border cases. Benchmark datasets don't have this problem, because they are 'cleaned', but in real life unsure cases often ...
3
votes
1answer
148 views

How does Mask R-CNN automatically output a different number of objects on the image?

Recently, I was reading Pytorch's official tutorial about Mask R-CNN. When I run the code on colab, it turned out that it automatically outputs a different number of channels during prediction. If the ...
3
votes
1answer
27 views

Is there a methodology for splitting up annotated orthophotos into smaller photos that retain the original bounding boxes?

I'm trying to train an object detection algorithm (i.e. YOLOv4 Scaled, Faster R-CNN) on data taken from large orthophotos. Let's say I have one class, and I label the entire orthophoto with bounding ...
3
votes
1answer
95 views

What is a heatmap in the CornerNet paper?

I have been working on understanding how CornerNet works, but I couldn't figure out a few parts about the architecture. First, the authors mention that there are 3 distinct parts to be predicted as a ...
3
votes
1answer
42 views

If two objects are too close to each other, would an object detector do a poor job of correctly classifying them?

Suppose we have an object detector that is trained to detect $20$ products. If two objects are too close to each other, in general, would an object detector do a poor job of correctly classifying them?...
3
votes
1answer
57 views

Do deeper residual networks perform better or worse?

If you have an $18$ layer residual network versus and a $32$ layer residual network, why would the former do better at object detection than the latter, if you have both models are training using the ...
3
votes
2answers
2k views

Calculation of FPS on object detection task

How to calculate mean speed in FPS for an object detection model like YOLOv3 or YOLOv3-Tiny? Different object detection models are often presented on charts like this: I am using the DarkNet ...
3
votes
1answer
32 views

Is it better to adjust the natural lighting (while recording the video) or to subsequently apply filters on the original video?

For the purpose of object detection, is it better to adjust the natural lighting (while recording the video) or to apply filters (e.g. brightness filters, etc.) on the original video to make it ...
3
votes
1answer
151 views

How to keep track of the same person detected in different frames of a camera?

At this moment, I am able to use NN to identify an object, such as a human, when given a frame from the camera. Once locate the object, then I can feed the human object image to either NN that's ...
3
votes
0answers
35 views

How are Ground truth provided to each Pyramid map in RetinaNet or YOLOv3 Paper? How is the mapping of Feature Pyramids done to Ground Truth

SO the YOLO V3 and RetinaNet both uses the Feature pyramids which look something like this: (except b and e which have one ...
3
votes
0answers
29 views

If random rotations are included in the data augmentation process, how are the new bounding boxes calculated?

When studying bounding box-based detectors, it's not clear to me if data augmentation includes adding random rotations. If random rotations are added, how is the new bounding box calculated?
3
votes
0answers
26 views

Defect Detection System using Deep Learning

What is the general approach to defect detection in deep learning? Would the approach be better if we try to learn the positive images (defects in images) as much as possible or we try to learn the ...
3
votes
0answers
93 views

Does Retina-net's focal loss accomplish its goal?

Taking out the weighting factor we can define focal loss as $$FL(p) = -(1-p)^\gamma log(p) $$ Where $p$ is the target probability. The idea being that single stage object detectors have a huge ...
3
votes
4answers
389 views

Can bounding boxes further improve the performance of a CNN classifier?

Suppose I have a standard image classification problem (i.e. CNN is shown a single image and predicts a single classification for it). If I were to use bounding boxes to surround the target image (i.e....
2
votes
3answers
3k views

Small size datasets for object detection, segmentation and localization [closed]

I am looking for a small size dataset on which I can implement object detection, object segmentation and object localization. Can anyone suggest me a dataset less than 5GB? Or do I need to know ...
2
votes
2answers
262 views

Which neural network is appropriate for measuring object dimensions from stereo images?

I have stereo pairs (left, right) images of concrete cracks. I want to measure the length of the crack from those image pairs. Which neural network is appropriate for measuring object dimensions from ...
2
votes
1answer
399 views

How to add negative samples for object detection?

My question is: how to add certain negative samples to the training dataset to suppress those samples that are recognized as the object. For example, if I want to train a car detector. All my training ...
2
votes
1answer
357 views

Why do we resize images before using them for object detection?

In object detection, we can resize images by keeping the ratio the same as the original image, which is often known as "letterbox" resize. My questions are Why do we need to resize images? ...
2
votes
1answer
33 views

How does a bounding box detection network "know" about absolute position?

I've always found bounding box regression a bit weird. There's no positional encoding like in vision transformers, so how does the network "know" the absolute position when producing ...
2
votes
1answer
48 views

Viola-Jones algorithm: Haar-like features, how are the features extracted?

If I have an image like this 1 2 3 4 5 6 7 8 a b c d e f g h ... And I apply a Haar-like feature with a template ...
2
votes
1answer
44 views

How does YOLO handle non-class objects?

I have been reading more about computer vision and I'm bothered by YOLO and similar deep learning architectures. The thing I am confused about is how non-class image sections are dealt with. In ...
2
votes
1answer
33 views

If an image contains two distinct objects, should I create a copy of this image with distinct labels for each copy?

Suppose we want to detect whether an object is one of the following classes: $\text{Object}_1, \text{Object}_2, \text{Object}_3$ and $\text{Person}$. Should the annotated images only contain bounding ...
2
votes
1answer
84 views

Is it possible to train a CNN to predict the dimensions of primitive objects from point clouds?

Is it possible to train a convolutional neural network (CNN) to predict the dimensions of primitive objects such as (spheres, cylinders, cuboids, etc.) from point clouds? The input to the CNN will be ...
2
votes
1answer
57 views

Why tf object detection api needs so few pictures?

I am wondering why tf object detection api needs so few picture samples for training while regular cnns needs many more? What I read in tutorials is that tf object detection api needs around 100-500 ...
2
votes
1answer
77 views

How should I detect an object in a camera image?

I would like to create a model, that will tell me if one type of object is in an image or not. So, for example, I have a camera and I would like to see when one object gets into the shot. Object ...
2
votes
1answer
108 views

What do we mean by 'principal angle between subspaces'?

I came across the term 'principal angle between subspaces' as a tool for comparing objects in images. All material that I found on the internet seems to deal with this idea in a highly mathematical ...
2
votes
2answers
436 views

Is the choice of the optimiser relevant when doing object detection?

Suppose that we have 4 types of dogs that we want to detect (Golden Retriever, Black Labrador, Cocker Spaniel, and Pit Bull). The training data consists of png images of a data set of dogs along with ...
2
votes
1answer
45 views

On which data evaluate an object detection model ? (similar or real life data ?)

I'm training an object detection model (SSD300) to detect and classify body poses in thermal images. Even I have more than 2k different poses, but the background does not change much (I have only 5 ...
2
votes
1answer
614 views

How does non-max suppression work when one or multiple bounding boxes are predicted for the same object?

My understanding of how non-max suppression works is that it suppresses all overlapping boxes that have a Jaccard overlap smaller than a threshold (e.g. 0.5). The boxes to be considered are on a ...
2
votes
1answer
53 views

Can a trained object detection model deal with variations of the input?

Suppose an object detection algorithm is good at detecting objects and people when an object and person is close to a camera and upright. If the person walks farther away from the camera and is "...
2
votes
1answer
81 views

How do I locate a specific object in an image?

Some pictures contain an elephant, others don't. I know which of the pictures contain the elephant, but I don't know where it is or how does it look like. How do I make a neural network which ...
2
votes
2answers
697 views

How to architect a network to find bounding boxes in simple images?

I have an application where I want to find the locations of objects on a simple, relatively constant background (fixed camera angle, etc). For investigative purposes, I've created a test dataset that ...
2
votes
2answers
169 views

Would YOLO be able to detect objects in "different" positions?

I have the following question about You Only Look Once (YOLO) algorithm, for object detection. I have to develop a neural network to recognize web components in web applications - for example, login ...
2
votes
2answers
2k views

How can I detect thin objects (like pens and pencils) without a bounding box but only 2 endpoints and the orientation?

I am looking to detect thin objects, like pens, pencils, and surgical instruments. The bounding box is not important, but I am looking to see if I can train a model to detect both the object as well ...
2
votes
1answer
130 views

Understanding a paragraph about object detection with two objects

I was reading this article on detecting rectangles in an image. My doubt is in the part where the model works fine with detecting a single object, but struggles with two rectangles detection. The ...
2
votes
0answers
25 views

What would be a reasonable option for clustering for unknown number of clusters and a lot of outliers?

I am implementing the CV detection pipeline with the use of SIFT and KNN Matcher. Image keypoints matched to the query keypoints produce the following image: The matched objects have a lot of key ...
2
votes
0answers
12 views

Can a GIoU loss (generalized intersection over union) be used after an STN module (spatial transformer network)?

I have a model that uses an STN module for number detection and Mean Squared Error loss. But I would like to replace it for GIoU, because MSE doesn't take into account how much of the target area has ...
2
votes
1answer
109 views

How do I choose the hyper-parameters for a model to detect different guitar chords?

I need to build a hand detector that recognizes the chord played by a hand on a guitar. I read this article Static Hand Gesture Recognition using Convolutional Neural Network with Data Augmentation ...
2
votes
0answers
40 views

Which method can accurately detect circular/angular shapes? (attached example)

Is there a method to detect shapes like these accurately and efficiently? I have tried the OpenCv Haar Casacade Classifier which does not work well. These shapes should all be the same class object ...