Analyzing Object Detection Dataset

In this tutorial, we will analyze an object detection dataset with bounding boxes and identify potential issues. We will cover the process of loading the annotations, visualizing bounding boxes, and finding objects that are possibly mislabeled or not useful for training (tiny, blurry, etc.).

When analyzing an object bounding box dataset, it is often useful to start with an image level analysis, to uncover low hanging fruits such as broken images and and duplicates. More on that in our Analyzing Image Classification tutorial.

In this tutorial we will analyze the image at the bounding box level.

Setting Up

You can follow along this tutorial by running this notebook on Google Colab

First, install fastdup with:

bash

pip install fastdup

And verify the installation.

Python

import fastdup
fastdup.__version__

This tutorial runs on version 0.906.

Download Dataset

We will be using the coco minitrain dataset for this tutorial.

coco-minitrainis a curated mini training set (25K images ≈ 20% of train2017) from the original COCO dataset.

Let's download the images and .csv annotations of the coco-minitrain dataset.

Shell

# Download images from mini-coco
!gdown --fuzzy https://drive.google.com/file/d/1iSXVTlkV1_DhdYpVDqsjlT4NJFQ7OkyK/view
!unzip coco_minitrain_25k.zip

# Download csv annotations
!cd coco_minitrain_25k/annotations && gdown --fuzzy https://drive.google.com/file/d/1i12p23cXlqp1QrXjAD_vu467r4q67Mq9/view

Load annotations

We will use a simple converter to convert the COCO format JSON annotation file into the fastdup annotation dataframe. This converter is applicable to any dataset which uses COCO format.

Python

import fastdup
import pandas as pd

coco_csv = 'coco_minitrain_25k/annotations/coco_minitrain2017.csv'
coco_annotations = pd.read_csv(coco_csv, header=None, names=['filename', 'col_x', 'row_y',
                                                             'width', 'height', 'label', 'ext'])

coco_annotations['split'] = 'train'  # Only train files were loaded
coco_annotations = coco_annotations.drop_duplicates()

Python

coco_annotations.head(3)

	img_filename	bbox_x	bbox_y	bbox_w	bbox_h	label	ext	split
0	000000131075.jpg	20	55	313	326	tv	0	train
1	000000131075.jpg	176	381	286	136	laptop	0	train
2	000000131075.jpg	369	361	72	73	laptop	0	train

Run fastdup

Python

image_dir = 'coco_minitrain_25k/images/train2017/'  # Train data only
work_dir = 'fastdup_coco_25k'
fd = fastdup.create(work_dir, image_dir)
fd.run(annotations=coco_annotations, overwrite=True)

FastDup Software, (C) copyright 2022 Dr. Amir Alush and Dr. Danny Bickson.
2023-02-26 11:34:40 [WARNING] Warning: test_dir and input_dir should not point to the same directory 
2023-02-26 11:34:40 [INFO] Going to loop over dir /var/folders/27/6tpv2d5j2yd9rtg1ml23f5h40000gn/T/tmpxnmh4kii.csv
2023-02-26 11:34:40 [INFO] Found total 183546 images to run on
[■■■■                                              ] 7% Estimated: 5 Minutes 0 Features

Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000402248.jpg. Please check bounding box file 14 241 3 0
Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000141426.jpg. Please check bounding box file 106 224 2 0
Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000564557.jpg. Please check bounding box file 30 271 5 0
Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000183338.jpg. Please check bounding box file 241 287 3 0
Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000088835.jpg. Please check bounding box file 292 200 3 0
Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000168890.jpg. Please check bounding box file 7 95 1 0
Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000171360.jpg. Please check bounding box file 339 181 8 0
Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000256896.jpg. Please check bounding box file 210 350 3 0
Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000389655.jpg. Please check bounding box file 372 438 3 0


2023-02-26 11:39:33 [ERROR] Error: found invalid bounding box for image /Users/amirm/vl_data/coco_minitrain_25k/images/train2017/000000389655.jpg. Please check bounding box file 372 438 3 0
2023-02-26 11:39:42 [INFO] Found total 183546 images to run onated: 0 Minutes 0 Features
2023-02-26 11:42:49 [INFO] 186133) Finished write_index() NN model
2023-02-26 11:42:49 [INFO] Stored nn model index file /Users/amirm/vl_workdir/coco_minitrain_25k/tutorial_bboxes/nnf.index
2023-02-26 11:43:18 [INFO] Total time took 517522 ms
2023-02-26 11:43:18 [INFO] Found a total of 163 fully identical images (d>0.990), which are 0.03 %
2023-02-26 11:43:18 [INFO] Found a total of 5895 nearly identical images(d>0.980), which are 1.07 %
2023-02-26 11:43:18 [INFO] Found a total of 93110 above threshold images (d>0.900), which are 16.91 %
2023-02-26 11:43:18 [INFO] Found a total of 16495 outlier images         (d<0.050), which are 3.00 %
2023-02-26 11:43:18 [INFO] Min distance found 0.483 max distance 1.000
2023-02-26 11:43:18 [INFO] Running connected components for ccthreshold 0.960000 

 ########################################################################################

Dataset Analysis Summary: 

    Dataset contains 183546 images
    Valid images are 89.87% (164,954) of the data, invalid are 10.13% (18,592) of the data
    For a detailed analysis, use `.invalids()`.

    Similarity:  3.96% (7,261) belong to 72 similarity clusters (components).
    96.04% (176,285) images do not belong to any similarity cluster.
    Largest cluster has 674 (0.37%) images.
    For a detailed analysis, use `.connected_components()`
(similarity threshold used is 0.9, connected component threshold used is 0.96).

    Outliers: 5.75% (10,553) of images are possible outliers, and fall in the bottom 5.00% of similarity values.
    For a detailed list of outliers, use `.outliers(data=True)`.

Get class statistics

Python

fd.annotations()['label'].value_counts()

person           50336
chair             7870
car               7703
bottle            4330
cup               4154
                 ...  
toothbrush         309
bear               296
parking meter      183
toaster             46
hair drier          40
Name: label, Length: 80, dtype: int64

Class distribution

The dataset contains 25k images and 183k objects, an average of 7.3 objects per image.

Interestingly, we see a highly unbalanced class distribution, where all 80 coco classes are present here, but there is a strong balance towards the person class, that accounts for over 56k instances (30.6%). Car and Chair classes also contain over 8k instances each, while at the bottom of the list the toaster and hair drier classes contain as few as 40 instances.

Using Plotly we get a useful interactive histogram.

Python

import plotly.express as px
fig = px.histogram(coco_annotations, x="label")
fig.show()

Find outliers and duplicates for specific classes

Similarity Clusters

First we visualize the general lists of duplicates and outliers, by default it is sorted by the number of elements:

Python

fd.vis.component_gallery()

Sorting by the largest objects

Python

fd.vis.component_gallery(metrix='size')

A few notable examples

Further down the gallery, a few examples of objects that might be erroneous or mislabeled.

1. In the red cup image, we see individual cups in the sleeve that are barely visible get a small bounding box that captures only the lip of the cup. For many cases, this kind of annotation is not useful.
2. In the cake image we can see the same object, annotated three times, each time with a different label (fork, knife and spoon) - in reality, it is very hard to tell what it actually is.

Showing galleries for a specific object class

Now we'll slice the gallery according specific classes - let's look at the 'person' class

Python

fd.vis.component_gallery(metrix='size', slice='person', max_width=900)

Visualizing outliers

Python

fd.vis.outliers_gallery()

Find size and shape issues

Objects come in various shapes and sizes, and some times objects might be incorrectly labeled or too small to be useful. We will now find the smallest, narrowest and widest objects, and asses their usefulness.

Python

df = fd.annotations()
df['area'] = df['width'] * df['height']
df['aspect'] = df['width'] / df['height']

Python

# Smallest 5% of objects:
smallest_objects = df[df['area'] < df['area'].quantile(0.05)].sort_values(by=['area'])
# 5% of extreme aspect ratios
aspect_ratio_objects = df[(df['aspect'] < df['aspect'].quantile(0.05))
                         |(df['aspect'] > df['aspect'].quantile(0.95))].sort_values(by=['aspect'])

Python

smallest_objects.head(3)

	img_filename	bbox_x	bbox_y	bbox_w	bbox_h	label	ext	split	fastdup_id	error_code	is_valid	area	aspect
fd_index
16444	000000535145.jpg	185	227	10	10	apple	0	train	16444	VALID	True	100	1.0
44855	000000553446.jpg	332	229	10	10	sports ball	0	train	44855	VALID	True	100	1.0
32115	000000283323.jpg	138	200	10	10	car	0	train	32115	VALID	True	100	1.0

Python

aspect_ratio_objects.head(3)

	img_filename	bbox_x	bbox_y	bbox_w	bbox_h	label	ext	split	fastdup_id	error_code	is_valid	area	aspect
fd_index
134869	000000093298.jpg	230	1	16	419	kite	0	train	134869	VALID	True	6704	0.038186
3642	000000002444.jpg	1	136	11	263	person	0	train	3642	VALID	True	2893	0.041825
164218	000000116502.jpg	222	63	16	318	spoon	0	train	164218	VALID	True	5088	0.050314

Python

aspect_ratio_objects.tail(3)

	img_filename	bbox_x	bbox_y	bbox_w	bbox_h	label	ext	split	fastdup_id	error_code	is_valid	area	aspect
fd_index
77844	000000444692.jpg	0	177	640	19	bench	0	train	77844	VALID	True	12160	33.684211
77850	000000444692.jpg	1	58	638	15	bench	0	train	77850	VALID	True	9570	42.533333
173220	000000516740.jpg	17	197	601	13	train	0	train	173220	VALID	True	7813	46.230769

Look at that! The slices reveal many items that are either tiny (10x10 pixels) or have extreme aspect ratios - as extreme at 1:46 - an object 601 pixels wide by only 13 pixels high.

Objects that didn't make the cut:

Let's look at objects deemed invalid by fastdup. These are either objects that are too small to be useful in our analysis (smaller than 10px), have bouding boxes with illeagal values (negative or beyond image boundaries), or are part of images that are missing. We can tell which is which by the error_code column in our dataframe.

Python

fd.invalid_instances().head(3)

	img_filename	bbox_x	bbox_y	bbox_w	bbox_h	label	ext	split	fastdup_id	error_code	is_valid
0	000000262162.jpg	437	244	19	9	mouse	0	train	16	ERROR_BAD_BOUNDING_BOX	False
1	000000524325.jpg	137	332	8	11	person	0	train	60	ERROR_BAD_BOUNDING_BOX	False
2	000000524325.jpg	177	294	5	11	person	0	train	65	ERROR_BAD_BOUNDING_BOX	False

Distribution of error codes:

A simple value_counts will tell us the distribution of the errors. We have found 18,592 (!) bounding boxes that are either too small or go beyond image boundaries. This is 10% of the data! Filtering them would both save us gruesome debugging of training errors and failures and help up provide the model with useful size objects.

Python

fd.invalid_instances().error_code.value_counts()

ERROR_BAD_BOUNDING_BOX    18592
Name: error_code, dtype: int64

Find possible mislabels

The fastdup similarity search and gallery is a strong tool for finding objects that are possibly mislabeled. By finding each object's nearest neighbors and their classes, we can find objects with classes contradicting their neighbors' - a strong sign for mislabels.

Let's visualize one of the more popular classes - 'chair':

Python

fd.vis.components_gallery(slice='diff')

There are many examples in the gallery, but let's look at the first few:

In the images above we see the many mislabeled 'chair' bounding boxes. There are multiple 'person' labeled as 'chair'.

This would be a good opportunity to go over the labeling definitions for both classes, and find cases where mix-ups can occur.

A next step could be a deeper dive into all images labeled as either chair or couch, with a lower value of clustering threshold (the ccthreshold parameter in .run()).

Summary

👍

We have shown multiple annotation issues for the mini-coco 25k image dataset

First, we looked at the class distribution, showing that some classes contain only a few dozen objects, which are not enough for reliable results.

Then we used fastdup to run an analysis and visualize both duplicate and outlier images, sorting according to various metrics - object size, similarity cluster image count, and mean cluster similarity score, surfacing all sorts of objects that could be removed to improve results.

Afterwards, we analyzed the bounding boxes for size and aspect ratio issues, and found over 18k object bounding boxes that are too small to be useful for most cases.

Finally, for finding mis-labels we looked into the most similar images in a specific class, and uncovered a common mixup between chair and couch classes. This is just a taste of fastdup label capabilities, which will be covered in further tutorials.

Looking for finding mislabels at scale? Sign up for fastdup enterprise.

GitHub • Join Slack Community • Discussion Forum Blog • Documentation • About Us LinkedIn • Twitter