Datasets:

HuggingFaceM4
/

LocalizedNarratives

Modalities:

Image

ArXiv:

License:

Dataset card Data Studio Files Files and versions

xet

Community

Dataset Preview

Duplicate

Subset (2)

OpenImages

Split (3)

train

The full dataset viewer is not available (click to read why). Only showing a preview of the rows.

Job manager was killed while running this job (job exceeded maximum duration).

Error code:   JobManagerExceededMaximumDurationError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

image_url string	dataset_id string	image_id string	annotator_id int32	caption string	timed_caption sequence	traces sequence	voice_recording string
https://s3.amazonaws.com…c17e61428d75.jpg	open_images	6375c17e61428d75	96	In this image there are some persons standing at right side of this image and there are some lights are arranged at top of this image. There are some cabins as we can see at left side of this image and there is a floor at bottom of this image.	{ "utterance": [ "In", "this image there are some persons", "standing", "at", "right", "side", "of", "this", "image", "and", "there are", "some", "lights", "are", "arranged", "at", "top of", "this", "image. There are some", "cabins", ...	[ { "x": [ 0.0869, 0.12, 0.1508, 0.1886, 0.2439, 0.3283, 0.3788, 0.4072, 0.419, 0.419, 0.4119, 0.3567, 0.337, 0.3038, 0.2912, 0.2904, 0.3014, 0.3283, 0.4293, 0.5034, 0.5342, 0.5436, ...	open_images_train/open_images_train_6375c17e61428d75_96.ogg
https://s3.amazonaws.com…d89eb7194b30.jpg	open_images	6d67d89eb7194b30	60	In this image we can see a pie on a large spatula. On the backside we can see some fire.	{ "utterance": [ "In", "this", "image", "we", "can", "see", "a", "pie", "on", "a", "large", "spatula.", "On", "the", "backside", "we", "can", "see", "some", "fire." ], "start_time": [ 0, 0.4, 0.5, 1, 1.5, 1.8, ...	[ { "x": [ 0.6955, 0.7084, 0.7176, 0.7203, 0.7203, 0.7203, 0.7084, 0.69, 0.6653, 0.635, 0.6047, 0.5744, 0.545, 0.5156, 0.4899, 0.466, 0.4596, 0.4559, 0.4559, 0.4642, 0.4761, 0.4954, ...	open_images_train/open_images_train_6d67d89eb7194b30_60.ogg
https://s3.amazonaws.com…9a7a89bacad0.jpg	open_images	01329a7a89bacad0	107	In this image, we can see a table, on that table there is a cake and we can see a candle on the cake, there is a black color laptop on the table, in the background we can see a wall and there are some photos on the wall.	{ "utterance": [ "In", "this image, we can see a", "table,", "on", "that", "table", "there", "is a", "cake", "and", "we", "can", "see", "a", "candle", "on", "the", "cake,", "there", "is", "a", "black", "color", "laptop", ...	[ { "x": [ 1.0512, 0.9031, 0.8367, 0.8014, 0.7903, 0.7837, 0.7792, 0.7682, 0.7505, 0.735, 0.7129, 0.693, 0.6775, 0.6554, 0.609, 0.5868, 0.5647, 0.5426, 0.5139, 0.4763, 0.4276, 0.3746...	open_images_train/open_images_train_01329a7a89bacad0_107.ogg
https://s3.amazonaws.com…1b9d50b4f7a5.jpg	open_images	c1cf1b9d50b4f7a5	5	In the image we can see a T-shirt, black in color, on it there is a text. We can even see there are many flower bookies. This is a footpath, brick wall and fruits.	{ "utterance": [ "In the", "image", "we", "can", "see", "a", "T-shirt,", "black", "in", "color,", "on", "it", "there is a", "text. We", "can", "even see there", "are", "many", "flower", "bookies.", "This is a footpath, brick wall", ...	[ { "x": [ -0.0872, 0.0299, 0.1086, 0.1799, 0.2439, 0.2897, 0.3153, 0.3171, 0.3098, 0.2842, 0.2531, 0.2366, 0.2329, 0.2878, 0.3372, 0.414, 0.5201, 0.6189, 0.6847, 0.7104, 0.7122, 0.6...	open_images_train/open_images_train_c1cf1b9d50b4f7a5_5.ogg
https://s3.amazonaws.com…95b2e8954792.jpg	open_images	a2b695b2e8954792	34	In this image in the foreground there are two persons one person is wearing a hat, and in the background there are group of people one person is holding a pole and flag and some boards poles, buildings, mountains. At the bottom there are some stones, and at the top there is sky.	{ "utterance": [ "In this", "image in", "the", "foreground", "there are two persons", "one", "person", "is", "wearing", "a", "hat,", "and", "in", "the", "background", "there", "are", "group", "of", "people", "one", "person", "...	[ { "x": [ 0.5953, 0.5572, 0.5417, 0.5417, 0.5488, 0.5601, 0.5953, 0.6869, 0.7193, 0.7432, 0.7616, 0.7771, 0.7897, 0.7996, 0.8081, 0.8165, 0.8207, 0.8236, 0.8278, 0.8292, 0.8236, 0.8...	open_images_train/open_images_train_a2b695b2e8954792_34.ogg
https://s3.amazonaws.com…f495671ea4d9.jpg	open_images	6932f495671ea4d9	1	In this picture we can see a wheel, a gear and springs in the front, in the background there is grass.	{ "utterance": [ "In", "this", "picture", "we", "can", "see", "a", "wheel, a", "gear", "and", "springs", "in", "the", "front, in the", "background", "there", "is", "grass." ], "start_time": [ 0, 0, 0.5, 0.7, 0.9, 1.1, ...	[ { "x": [ 0.3916, 0.4567, 0.5621, 0.6652, 0.7072, 0.8135, 0.8801, 0.8999, 0.9015, 0.9103, 0.9158, 0.9229, 0.9317, 0.9372, 0.9396, 0.9428, 0.9451, 0.9459, 0.9459, 0.9459, 0.9459, 0.9...	open_images_train/open_images_train_6932f495671ea4d9_1.ogg
https://s3.amazonaws.com…2763a7fe8f66.jpg	open_images	bbae2763a7fe8f66	27	Here we can see a woman having a football in her hand and behind her we can see group of people standing and it is snowy	{ "utterance": [ "Here", "we", "can", "see", "a", "woman", "having", "a", "football", "in", "her", "hand", "and", "behind", "her", "we", "can", "see", "group", "of", "people", "standing", "and", "it", "is", "snowy"...	[ { "x": [ 0.4026, 0.4219, 0.4292, 0.4393, 0.4577, 0.466, 0.4752, 0.4779, 0.4834, 0.4862, 0.4871, 0.489, 0.488, 0.488, 0.4899, 0.4945, 0.5018, 0.5028, 0.5009, 0.4991, 0.4917, 0.4853,...	open_images_train/open_images_train_bbae2763a7fe8f66_27.ogg
https://s3.amazonaws.com…a5c5fd0b4768.jpg	open_images	75faa5c5fd0b4768	21	In this image there is a bed ,on the bed there are the pillows and beside the bed there is a table ,on the table there is a laptop and there is a fish pot ,under the table there is a bottle kept on the table and there is a speaker on the floor.	{ "utterance": [ "In", "this image there is a", "bed", ",on", "the", "bed there are the pillows and beside the", "bed", "there", "is", "a", "table", ",on", "the", "table", "there is a", "laptop", "and there", "is", "a", "fish", "pot",...	[ { "x": [ 0.0348, 0.0413, 0.0505, 0.0587, 0.0642, 0.0688, 0.0734, 0.078, 0.0835, 0.0891, 0.0983, 0.1056, 0.1157, 0.1212, 0.1277, 0.1341, 0.1405, 0.147, 0.1525, 0.1589, 0.1663, 0.173...	open_images_train/open_images_train_75faa5c5fd0b4768_21.ogg
https://s3.amazonaws.com…eca320e7bd78.jpg	open_images	39eaeca320e7bd78	77	In this picture we can see vehicles on the road, trees, poles and in the background we can see the sky with clouds.	{ "utterance": [ "In", "this", "picture", "we", "can", "see", "vehicles", "on", "the", "road,", "trees,", "poles", "and", "in", "the", "background we can see the", "sky", "with", "clouds." ], "start_time": [ 0.4, 0.9, 1, 1...	[ { "x": [ 0.136, 0.136, 0.1388, 0.1426, 0.1444, 0.1491, 0.1593, 0.1733, 0.1817, 0.191, 0.2003, 0.2142, 0.2291, 0.2468, 0.2654, 0.285, 0.3101, 0.3343, 0.3613, 0.4004, 0.4255, 0.4488,...	open_images_train/open_images_train_39eaeca320e7bd78_77.ogg
https://s3.amazonaws.com…c810859998c8.jpg	open_images	6ae2c810859998c8	108	In this picture, we can see a few buildings, poles, stairs, railing, road, a few people, and vehicles, and the sky with clouds, and we can see date and time in bottom right side of the picture.	{ "utterance": [ "In", "this", "picture,", "we", "can", "see a few", "buildings,", "poles,", "stairs,", "railing,", "road, a", "few", "people, and vehicles,", "and", "the", "sky", "with", "clouds,", "and", "we can see", "date", "a...	[ { "x": [ 0.3466, 0.3552, 0.3617, 0.3617, 0.3704, 0.3876, 0.4125, 0.4395, 0.473, 0.5292, 0.5799, 0.6242, 0.6459, 0.6459, 0.6437, 0.607, 0.5335, 0.4698, 0.4352, 0.42, 0.4179, 0.4244,...	open_images_train/open_images_train_6ae2c810859998c8_108.ogg
https://s3.amazonaws.com…02ac1436cbfc.jpg	open_images	0b4002ac1436cbfc	46	This image is taken in a store where we can see mirrors, chairs, wall, lights, ceiling, bottles, tripod stands, frames, jar, table and a cupboard.	{ "utterance": [ "This image is", "taken", "in", "a", "store", "where", "we", "can", "see", "mirrors,", "chairs, wall,", "lights,", "ceiling,", "bottles,", "tripod", "stands,", "frames,", "jar,", "table", "and a", "cupboard." ], "...	[ { "x": [ 0.4028, 0.4244, 0.4429, 0.4599, 0.4784, 0.4969, 0.517, 0.5386, 0.5555, 0.5833, 0.6172, 0.6296, 0.6435, 0.6589, 0.6712, 0.6805, 0.6897, 0.6928, 0.7006, 0.7083, 0.7098, 0.71...	open_images_train/open_images_train_0b4002ac1436cbfc_46.ogg

End of preview.

Dataset Card for [Dataset Name]

Dataset Summary

Localized Narratives, a new form of multimodal image annotations connecting vision and language. We ask annotators to describe an image with their voice while simultaneously hovering their mouse over the region they are describing. Since the voice and the mouse pointer are synchronized, we can localize every single word in the description. This dense visual grounding takes the form of a mouse trace segment per word and is unique to our data. We annotated 849k images with Localized Narratives: the whole COCO, Flickr30k, and ADE20K datasets, and 671k images of Open Images, all of which we make publicly available.

As of now, there is only the OpenImages subset, but feel free to contribute the other subset of Localized Narratives!

OpenImages_captions is similar to the OpenImages subset. The differences are that captions are groupped per image (images can have multiple captions). For this subset, timed_caption, traces and voice_recording are not available.

Supported Tasks and Leaderboards

[More Information Needed]

Languages

[More Information Needed]

Dataset Structure

Data Instances

Each instance has the following structure:

{
  dataset_id: 'mscoco_val2017',
  image_id: '137576',
  annotator_id: 93,
  caption: 'In this image there are group of cows standing and eating th...',
  timed_caption: [{'utterance': 'In this', 'start_time': 0.0, 'end_time': 0.4}, ...],
  traces: [[{'x': 0.2086, 'y': -0.0533, 't': 0.022}, ...], ...],
  voice_recording: 'coco_val/coco_val_137576_93.ogg'
}

Data Fields

Each line represents one Localized Narrative annotation on one image by one annotator and has the following fields:

dataset_id: String identifying the dataset and split where the image belongs, e.g. mscoco_val2017.
image_id String identifier of the image, as specified on each dataset.
annotator_id Integer number uniquely identifying each annotator.
caption Image caption as a string of characters.
timed_caption List of timed utterances, i.e. {utterance, start_time, end_time} where utterance is a word (or group of words) and (start_time, end_time) is the time during which it was spoken, with respect to the start of the recording.
traces List of trace segments, one between each time the mouse pointer enters the image and goes away from it. Each trace segment is represented as a list of timed points, i.e. {x, y, t}, where x and y are the normalized image coordinates (with origin at the top-left corner of the image) and t is the time in seconds since the start of the recording. Please note that the coordinates can go a bit beyond the image, i.e. <0 or >1, as we recorded the mouse traces including a small band around the image.
voice_recording Relative URL path with respect to https://storage.googleapis.com/localized-narratives/voice-recordings where to find the voice recording (in OGG format) for that particular image.