Datasets:

nikos74
/

BMRS

task_id string	task string	reference_answer string	correct_examples list	incorrect_examples list	task_type string	complexity string
bb_m_01	What property distinguishes the right class from the left class?	shore as a smooth line \| shore as a jagged line	[ "smooth coastline \| ribbed coastline" ]	[ "\"regular\" shapes \| irregular shapes" ]	shape	2
bb_m_02	What property distinguishes the right class from the left class?	elongated object \| non-elongated object	[ "elongated objects \| non elongated objects", "longer \| shorter" ]	[ "Rectangular objects \| Round objects" ]	shape	1
bb_m_03	What property distinguishes the right class from the left class?	object boundary fits into frame fully \| object boundaries doesn't fit into frame	[ "object (island or lake) fits into the frame fully \| no islands or lakes present fully in the image" ]	[ "there is a small island in each picture \| most of the pictures are occupied by greenery" ]	shape	2
bb_m_04	What property distinguishes the right class from the left class?	vehicles present \| there are no vehicles	[ "cars present \| immovable objects" ]	[ "parking lots \| land plot" ]	presence	1
bb_m_08	What property distinguishes the right class from the left class?	all cars are parked parallel \| cars parked at different angles	[ "Orientation of cars along one straight line \| Cars are not directed along one straight line", "cars parallel \| cars not parallel" ]	[ "enlarged photograph, car parking, vehicles \| reduced photograph, car parking, vehicles", "car parking \| road" ]	spatial	3
bb_m_09	What property distinguishes the right class from the left class?	continuous curved lines \| branched lines	[ "Solid curved lines \| Branched lines" ]	[ "Solid curved lines \| Branched lines" ]	shape	1
bb_m_10	What property distinguishes the right class from the left class?	farm field \| nature	[ "Artificially marked land plots \| Natural irregularly shaped landscapes" ]	[ "field \| forest" ]	semantic	1
bb_m_100	What property distinguishes the right class from the left class?	a road \| an intersection	[ "Line \| Cross" ]	[ "top-down view \| angled view" ]	shape	1
bb_m_101	What property distinguishes the right class from the left class?	vertical patterns on the sand \| diagonal patterns on the sand	[ "texture stripes are located approximately vertically \| texture stripes are located at different angles" ]	[ "Solid rock \| Sand" ]	spatial	1
bb_m_102	What property distinguishes the right class from the left class?	vertical lines \| horizontal lines	[ "scene objects form vertical lines \| scene objects form horizontal lines" ]	[ "Field \| Forested area" ]	spatial	2
bb_m_103	What property distinguishes the right class from the left class?	a baseball field in the lower left corner \| a baseball field in the upper right corner	[ "Orientation diagonally upward to the right \| Orientation diagonally downward to the left" ]	[ "arbitrary baseball field \| even baseball field" ]	spatial	1
bb_m_104	What property distinguishes the right class from the left class?	a turning road \| a straight road	[ "There is an elongated curve \| There is an elongated straight line" ]	[ "Houses predominately \| Nature" ]	shape	1
bb_m_105	What property distinguishes the right class from the left class?	the forest boundary is convex \| the forest boundary is concave	[ "Pieces of forest among open spaces \| Open spaces among forest" ]	[ "more natural colors \| less natural colors" ]	shape	2
bb_m_106	What property distinguishes the right class from the left class?	the coast boundary is convex \| the coast boundary is concave	[ "land is surrounded by water \| water is surrounded by land" ]	[ "wild beach \| not wild beach" ]	shape	2
bb_m_107	What property distinguishes the right class from the left class?	sandy beach \| rocky shore	[ "more even shore \| uneven shore" ]	[ "Even coastlines, the overall shade of the photographs is green \| Curved/winding coastlines, the overall shade of the photographs is blue" ]	semantic	1
bb_m_108	What property distinguishes the right class from the left class?	the coast is rounded \| the coast is straight	[ "complex coastline \| even coastline" ]	[ "smooth separation boundaries \| Straight separation boundaries" ]	shape	1
bb_m_109	What property distinguishes the right class from the left class?	urbanized river bank \| river bank with greenery	[ "Coast of an urbanized agglomeration \| Coast of a suburb or uninhabited territories", "bridges \| no bridges" ]	[ "river \| lake" ]	semantic	3
bb_m_11	What property distinguishes the right class from the left class?	city \| nature	[ "Buildings \| Wild area" ]	[ "No trees \| Trees" ]	semantic	1
bb_m_110	What property distinguishes the right class from the left class?	intersection/interchange in the city \| intersection/interchange outside the city	[ "city \| steppe" ]	[ "Circular square \| Ring road" ]	semantic	2
bb_m_111	What property distinguishes the right class from the left class?	racing track \| interchange	[ "special tracks \| real road interchanges" ]	[ "Country highway \| City highway" ]	semantic	2
bb_m_112	What property distinguishes the right class from the left class?	pool above the house \| pool below the house	[ "Pools are located on the northern side \| Pools are located on the southern side" ]	[ "green is represented in a narrow dynamic color range \| green is represented in a wide dynamic color range" ]	spatial	2
bb_m_113	What property distinguishes the right class from the left class?	rectangle on the roof \| circle on the roof	[ "Object with a quadrilateral in the center \| Object with a circle in the center" ]	[ "Pronounced inner cutout in the roofs of the central buildings \| Solid roofs on the central buildings" ]	shape	2
bb_m_114	What property distinguishes the right class from the left class?	rivers \| roads	[ "Naturally created road \| Road artificially created by humans" ]	[ "Natural landscapes \| Artificial landscapes" ]	semantic	1
bb_m_115	What property distinguishes the right class from the left class?	power transmission line tower \| wind power plant	[ "Power transmission towers \| Wind power plants" ]	[ "no shadows \| shadows present" ]	semantic	1
bb_m_116	What property distinguishes the right class from the left class?	sheep are small \| sheep are large	[ "Animals in the distance \| Animals nearby" ]	[ "Tree shadows are cast onto the pasture, sheep shadows are not distinguishable \| Sheep shadows are distinguishable and clearly visible" ]	size	1
bb_m_117	What property distinguishes the right class from the left class?	city \| animals	[ "Buildings \| Nature" ]	[ "pond \| ground" ]	semantic	1
bb_m_118	What property distinguishes the right class from the left class?	sheep in one cluster \| sheep in several clusters	[ "grouped in one place \| not grouped in one place" ]	[ "sheep \| goats" ]	number	2
bb_m_12	What property distinguishes the right class from the left class?	water present \| water absent	[ "Water \| Forests" ]	[ "Nature \| fields" ]	presence	1
bb_m_120	What property distinguishes the right class from the left class?	only passenger cars or only trucks \| both passenger cars and trucks	[ "Same-type vehicles \| Vehicles of different types" ]	[ "several cars \| trucks" ]	same	3
bb_m_121	What property distinguishes the right class from the left class?	presence of haze, clouds \| absence of haze, clouds	[ "presence of clouds \| absence of clouds" ]	[ "Urbanised area, lots of clouds \| Rural area, few clouds" ]	semantic	1
bb_m_122	What property distinguishes the right class from the left class?	camels \| zebras	[ "Camels \| Non-camels" ]	[ "desert \| savanna" ]	semantic	2
bb_m_124	What property distinguishes the right class from the left class?	inanimate objects \| animals	[ "Transport \| Nature" ]	[ "Planes \| a sheep" ]	semantic	1
bb_m_125	What property distinguishes the right class from the left class?	animals of one color \| animals of different colors	[ "all individuals are identical \| there are distinct individuals" ]	[ "animals together \| animals running apart" ]	same	2
bb_m_126	What property distinguishes the right class from the left class?	3d render \| real imaging	[ "Images from videogames \| Real images", "3d graphics \| remote sensing imagery" ]	[ "One \"type\" of surface touches the border of the image (whether water or land) \| Two \"types\" of surface touch the border of the image (land and forest, water and land)" ]	semantic	1
bb_m_13	What property distinguishes the right class from the left class?	absence of objects \| presence of objects	[ "areas without objects \| areas with objects", "empty \| objects present" ]	[ "No transport \| There is transport" ]	presence	1
bb_m_14	What property distinguishes the right class from the left class?	squares \| circles	[ "Quadrilaterals \| Circles" ]	[ "quadrilaterals \| triangles" ]	shape	1
bb_m_15	What property distinguishes the right class from the left class?	presence of bridges \| absence of bridges	[ "there are line segments \| there are no line segments" ]	[ "more water \| less water" ]	presence	1
bb_m_18	What property distinguishes the right class from the left class?	artificial reservoirs \| natural reservoirs	[ "Manmade \| Nature" ]	[ "regular shapes \| irregular shapes" ]	semantic	2
bb_m_19	What property distinguishes the right class from the left class?	presence of haze, clouds \| absence of haze, clouds	[ "Cloudy \| Clear" ]	[ "Big city, more clouds \| Suburb, less clouds" ]	presence	1
bb_m_20	What property distinguishes the right class from the left class?	forests \| fields	[ "there are trees \| there are no trees" ]	[ "horizontal lines \| vertical lines" ]	semantic	1
bb_m_22	What property distinguishes the right class from the left class?	vertical roads \| horizontal roads	[ "Vertical straight lines \| Horizontal straight lines" ]	[ "Highway \| main road" ]	spatial	1
bb_m_23	What property distinguishes the right class from the left class?	colored objects \| there are no colored objects	[ "Non-white objects \| White images" ]	[ "fewer lines \| more lines" ]	semantic	1
bb_m_24	What property distinguishes the right class from the left class?	twists clockwise from the center \| twists counterclockwise from the center	[ "vortex twisting to the right \| vortex twisting to the left", "Northern Hemisphere \| Southern Hemisphere" ]	[ "vortex \| spiral" ]	spatial	2
bb_m_27	What property distinguishes the right class from the left class?	presence of snow \| absence of snow	[ "Cold region \| warm region", "winter \| summer", "snow present \| snow absent" ]	[ "Mountain landscape \| Plain landscape" ]	semantic	2
bb_m_28	What property distinguishes the right class from the left class?	green trees \| bare trees	[ "there are leaves \| there are no leaves" ]	[ "Summer \| winter" ]	semantic	2
bb_m_29	What property distinguishes the right class from the left class?	four roads \| three roads	[ "Interchange for four roads \| Interchange for three roads" ]	[ "intechange \| road crossing" ]	number	1
bb_m_30	What property distinguishes the right class from the left class?	round objects \| square objects	[ "Circles \| Squares", "circular objects \| Rectangular objects" ]	[ "Solid curved lines \| Branched lines" ]	shape	1
bb_m_32	What property distinguishes the right class from the left class?	left-hand traffic \| right-hand traffic	[ "left-hand traffic \| right-hand traffic" ]	[ "agricultural fields \| natural objects" ]	spatial	3
bb_m_33	What property distinguishes the right class from the left class?	baseball field \| tennis court	[ "Baseball field \| Tennis court" ]	[ "Road without turns \| Intersection" ]	semantic	1
bb_m_36	What property distinguishes the right class from the left class?	plowed field, without greenery \| seeded field	[ "Freshly plowed land, little vegetation \| Field has sprouted, lots of vegetation" ]	[ "chaos \| orderliness" ]	semantic	2
bb_m_37	What property distinguishes the right class from the left class?	mountains without snow \| mountains with snow	[ "no snow \| there is snow" ]	[ "airplanes \| private houses" ]	presence	1
bb_m_38	What property distinguishes the right class from the left class?	wavy sea with foam \| calm sea without foam	[ "Waves \| Calm" ]	[ "yellow markings \| white markings" ]	semantic	1
bb_m_41	What property distinguishes the right class from the left class?	pool surrounded by houses \| pool not surrounded by houses	[ "pool surrounded by buildings on all sides \| pool not surrounded by buildings on all sides" ]	[ "More than 1 boat \| 1 boat" ]	spatial	3
bb_m_45	What property distinguishes the right class from the left class?	sports fields \| agricultural fields	[ "sports fields \| fields" ]	[ "Most likely large scale \| Most likely small scale" ]	semantic	1
bb_m_46	What property distinguishes the right class from the left class?	top-down view \| 45 degree view	[ "Top view shot \| Angled shot" ]	[ "Low-rise houses \| High-rise buildings" ]	spatial	3
bb_m_49	What property distinguishes the right class from the left class?	bridge \| road	[ "bridge \| not a bridge" ]	[ "2 standout objects \| 3 contrastively standout objects" ]	semantic	1
bb_m_51	What property distinguishes the right class from the left class?	boats are moving \| boats are stationary	[ "Moving vessels \| Vessels at parking" ]	[ "Interchange with circles \| Interchange without circles" ]	semantic	2
bb_m_53	What property distinguishes the right class from the left class?	A yellow marking in the center of the road \| A white marking only in the center of the road	[ "central road markings are yellow \| central road markings are white" ]	[ "Cars are located parallel to each other \| Cars are located diagonally (perpendicularly)" ]	semantic	2
bb_m_54	What property distinguishes the right class from the left class?	straight pedestrian crossing \| diagonal crossing	[ "\"zebra\" crossings are even rectangles \| zebra crossings\" are represented as beveled (not at 90 degrees)" ]	[ "Pedestrian crossings \| Pedestrian crossings" ]	shape	2
bb_m_56	What property distinguishes the right class from the left class?	dashed lines along the edges of the road \| solid lines along the edges of the road	[ "roadside is marked with a \"dashed\" line \| roadside is marked with a solid line (or not marked)" ]	[ "single-lane traffic \| two-lane traffic" ]	shape	3
bb_m_57	What property distinguishes the right class from the left class?	dashed lines in the middle of the road \| solid lines in the middle of the road	[ "dashed \| solid" ]	[ "Single line without branches \| All lines have branches" ]	shape	2
bb_m_58	What property distinguishes the right class from the left class?	houses are not enclosed \| enclosed houses	[ "Incomplete figures \| Complete figures" ]	[ "I don't know \| I don't know" ]	shape	2
bb_m_59	What property distinguishes the right class from the left class?	forest \| park	[ "Wild forest \| Forest in the city" ]	[ "Larger scale. Fewer buildings. \| Smaller scale. More buildings." ]	semantic	1
bb_m_60	What property distinguishes the right class from the left class?	pool \| lake	[ "pool \| lake, closed reservoir" ]	[ "boundaries of water bodies have predominantly angular contours \| boundaries of water bodies have predominantly rounded contours" ]	semantic	1
bb_m_61	What property distinguishes the right class from the left class?	straight highway \| interchange	[ "Straight road \| Circular road" ]	[ "Roads are visible in detail \| Roads are visible small" ]	shape	1
bb_m_62	What property distinguishes the right class from the left class?	highway \| railway	[ "road \| rails" ]	[ "Several boats \| One boat" ]	semantic	1
bb_m_63	What property distinguishes the right class from the left class?	the nose of the airplane is directed upward \| the nose of the airplane is directed to the right	[ "vertical airplane shots \| horizontal airplane shots" ]	[ "looks like computer graphics \| looks like real photographs" ]	spatial	1
bb_m_65	What property distinguishes the right class from the left class?	all roofs are of the same color \| there are roofs of different colors	[ "Roofs of houses of the same color \| Roofs of houses of different colors" ]	[ "ruler-like layout \| chaotically built buildings" ]	same	1
bb_m_66	What property distinguishes the right class from the left class?	objects are directed in different directions \| objects are directed in one direction	[ "chaotic form of vehicle parking \| all vehicles are lined up" ]	[ "different means of transport \| means of transport are grouped" ]	same	3
bb_m_67	What property distinguishes the right class from the left class?	straight road \| turning road	[ "Straight roads \| Curved roads" ]	[ "circle in the city \| circle in the field" ]	shape	1
bb_m_68	What property distinguishes the right class from the left class?	multi-lane road \| single-lane road	[ "many lanes \| 2 lanes" ]	[ "Airplanes on the runway \| Residential houses" ]	number	2
bb_m_69	What property distinguishes the right class from the left class?	boundary of three colors \| boundary of two colors	[ "three colors \| two colors" ]	[ "animals of the same color \| animals of different colors" ]	number	2
bb_m_70	What property distinguishes the right class from the left class?	cloverleaf interchange \| diamond interchange	[ "presence of four circular objects \| objects intersect each other with straight lines" ]	[ "Animals in the distance \| Animals nearby" ]	shape	1
bb_m_71	What property distinguishes the right class from the left class?	cross intersection \| roundabout	[ "intersections \| circles" ]	[ "Symmetrically \| Asymmetrically" ]	shape	1
bb_m_73	What property distinguishes the right class from the left class?	railway bridge over cars \| car bridge over the railway	[ "Railway \| Road" ]	[ "Industrial district of the city \| City bridges" ]	spatial	3
bb_m_74	What property distinguishes the right class from the left class?	the bridge over the road is vertical \| the bridge over the road is horizontal	[ "The overpass goes from the bottom border to the top border of the image \| The overpass goes from the left border to the right border of the image" ]	[ "More than one vehicle \| No more than one vehicle" ]	spatial	3
bb_m_75	What property distinguishes the right class from the left class?	the river is straight \| the river turns	[ "The river crosses the frame relatively straight \| The river makes at least 1 turn greater than 30 degrees" ]	[ "Road intersections \| Roundabout intersections" ]	shape	1
bb_m_76	What property distinguishes the right class from the left class?	city block development \| radial development	[ "Grid-like, square-nested structure \| Structure with radial distribution" ]	[ "Area with reservoirs \| Area without reservoirs" ]	shape	2
bb_m_79	What property distinguishes the right class from the left class?	island \| peninsula	[ "island \| peninsula", "coastline is closed within frame \| coasstline crosses the frame" ]	[ "Smooth coastline \| Broken coastline" ]	shape	1
bb_m_80	What property distinguishes the right class from the left class?	mountains \| quarries	[ "natural irregularities \| artificial irregularities" ]	[ "uniform landscape, especially in color \| diverse landscape in color" ]	semantic	3
bb_m_83	What property distinguishes the right class from the left class?	real image \| map image	[ "Satellite image \| Schematic map", "a real photo \| not a real photo" ]	[ "mountains \| quarry" ]	semantic	1
bb_m_87	What property distinguishes the right class from the left class?	airport \| railway station	[ "Airports \| Railway stations" ]	[ "there are areas with rounded corners \| there are areas with sharp corners" ]	semantic	3
bb_m_89	What property distinguishes the right class from the left class?	symmetrical interchange \| non-symmetrical interchange	[ "Symmetrical interchange system \| Asymmetrical interchange system" ]	[ "4 \"circles\" \| 2 \"circles\"" ]	shape	3
bb_m_91	What property distinguishes the right class from the left class?	roundabout with a bridge \| roundabout without a bridge	[ "roads intersect the circle \| roads do not intersect the circle" ]	[ "highway \| roundabout, circle" ]	shape	2
bb_m_92	What property distinguishes the right class from the left class?	the interchange bridge is vertical \| the interchange bridge is horizontal	[ "The horizontal road passes under the vertical one. \| The vertical road passes under the horizontal one" ]	[ "circular interchanges \| bridges" ]	spatial	3
bb_m_94	What property distinguishes the right class from the left class?	port without ships \| port with ships	[ "Without ships \| With ships" ]	[ "base near water \| parking on water for yachts" ]	presence	1
bb_m_95	What property distinguishes the right class from the left class?	car to the left of the house \| car to the right of the house	[ "Cars on the left side of the buildings \| Cars on the right side of the buildings" ]	[ "there are clouds \| there are no clouds" ]	spatial	3
bb_m_96	What property distinguishes the right class from the left class?	forest at the top \| forest at the bottom	[ "Trees in the upper part \| trees in the lower part" ]	[ "Completely forest territories \| There are residential territories" ]	spatial	2
bb_m_97	What property distinguishes the right class from the left class?	red object on the left \| red object on the right	[ "Bright red objects are located predominantly in the left part/half of the image. \| Bright red objects are located predominantly in the right part/half of the image." ]	[ "multi-lane road \| two-lane road" ]	spatial	2
bb_m_98	What property distinguishes the right class from the left class?	diagonal of the coastline from the lower right corner to the upper left \| diagonal of the coastline from the lower left to the upper right	[ "The shore stretches from the upper left corner to the lower right \| The shore stretches from the lower left corner to the upper right" ]	[ "Ordinary road intersections \| Roundabout intersections" ]	spatial	2
bb_r_38	What property distinguishes the right class from the left class?	lakes \| body of water not closed within the frame	[ "The reservoir is enclosed \| The reservoir is not bounded by the image borders on at least one side" ]	[ "Dark water \| Light water" ]	semantic	2
bb_r_48	What property distinguishes the right class from the left class?	high-rise buildings \| low-rise buildings	[ "multi-story buildings over 10 floors \| low-rise houses" ]	[ "rounded boundaries \| straight boundaries" ]	size	1
bb_r_99	What property distinguishes the right class from the left class?	inside of the roundabout is empty \| there is something inside the roundabout	[ "the center of the circle is empty \| non-empty center of the circle" ]	[ "Roundabout traffic \| Roundabout traffic" ]	presence	2
bb_s_05	What property distinguishes the right class from the left class?	many people \| one person	[ "Several people \| one person" ]	[ "All animals in the image are of the same color \| Each image contains animals of at least 2 colors" ]	number	1
bb_s_08	What property distinguishes the right class from the left class?	cars on the left \| cars on the right	[ "Cars on the left \| Cars on the right" ]	[ "Organized herd \| Chaotic herd" ]	spatial	2
bb_s_11	What property distinguishes the right class from the left class?	many cars \| one car	[ "There are several cars in the image \| There is only one car in the image" ]	[ "there is parking for cars \| there is no parking for cars" ]	number	1
bb_s_12	What property distinguishes the right class from the left class?	cars vertically \| cars horizontally	[ "Cars are visually positioned vertically \| Cars are visually positioned horizontally" ]	[ "land areas with reservoirs \| land areas without reservoirs" ]	spatial	1
bb_s_14	What property distinguishes the right class from the left class?	ships on the left \| ships on the right	[ "Transport in the left part of the image \| Transport in the right part of the image" ]	[ "boat \| boat" ]	spatial	2
bb_s_15	What property distinguishes the right class from the left class?	ships at the top \| ships at the bottom	[ "Transport in the upper part of the image \| Transport in the lower part of the image" ]	[ "ship \| ship" ]	spatial	2
bb_s_17	What property distinguishes the right class from the left class?	many ships \| one ship	[ "Many ships \| Few ships" ]	[ "Pier \| Boat afloat" ]	number	1

End of preview. Expand in Data Studio

BMRS: Bongard-Maximov Problems for Remote Sensing

Official Dataset for BMRS: Bongard–Maximov Problems for Remote Sensing Preprints.org
Official Code: GitHub.

This dataset includes:

BMRS/

Bongard problems are stored as complete collages of images and separate images. For each Bongard Problem:

right/ - all right images
left/ - all left images
pairs/ - image pairs
collage.png - complete problem
left.png - left side collage
right.png - right side collage

/LLM-as-a-Judge

Additional data needed to run the benchmark.

answer_examples_en.json - correct and incorrect answers examples
judge_system_prompt_en.txt - base prompt for the judge
reference_answers_en.json - reference correct answers
task_type_judge_prompts_en.json - task type specific prompts for judge
tasks_en.json - task types list for each problem

Dataset Structure


BMRS/
├── BMRS/
│   ├── bb_m_01/
│       ├── left/
│       │   ├── im0.png
│       │   ├── ...
│       │   └── im5.png
│       ├── pairs/
│       │   ├── im0.png
│       │   ├── ...
│       │   └── im5.png
│       ├── right/
│       │   ├── im0.png
│       │   ├── ...
│       │   └── im5.png
│       ├── collage.png
│       ├── left.png
│       └── right.png
│   ├── ...
│   └── bb_s_59/
│       ├── left/
│       │   ├── im0.png
│       │   ├── ...
│       │   └── im5.png
│       ├── pairs/
│       │   ├── im0.png
│       │   ├── ...
│       │   └── im5.png
│       ├── right/
│       │   ├── im0.png
│       │   ├── ...
│       │   └── im5.png
│       ├── collage.png
│       ├── left.png
│       └── right.png
├── LLM-as-a-Judge/
│   ├── answer_examples_en.json
│   ├── judge_system_prompt_en.txt
│   ├── reference_answers_en.json
│   ├── task_type_judge_prompts_en.json
│   └── tasks_en.json
├── data/
    └── test.parquet
prompting_strategies.json

Changelog

2026-10-29

Added Google Earth to the Data Source section.
Added attribution notes documenting the loss of original imagery metadata and the inability to identify imagery providers for individual samples.
Clarified licensing conditions for third-party imagery, including Google Earth.

Data Source

AID: Xia, Gui-Song, et al. "AID: A benchmark data set for performance evaluation of aerial scene classification." IEEE Transactions on Geoscience and Remote Sensing 55.7 (2017): 3965-3981.
VisDrone: Zhu, Pengfei, et al. "Detection and tracking meet drones challenge." IEEE transactions on pattern analysis and machine intelligence 44.11 (2021): 7380-7399.
DOTA: Ding, Jian, et al. "Object detection in aerial images: A large-scale benchmark and challenges." IEEE transactions on pattern analysis and machine intelligence 44.11 (2021): 7778-7796.
MLRSNet: Qi, Xiaoman, et al. "MLRSNet: A multi-label high spatial resolution remote sensing dataset for semantic scene understanding." ISPRS Journal of Photogrammetry and Remote Sensing 169 (2020): 337-350.
HyperHazeOff: Nikonorov, Artem, et al. "HyperHazeOff: Hyperspectral Remote Sensing Image Dehazing Benchmark." Journal of Imaging 11.12 (2025): 422.
WAID: Mou, Chao, et al. "Waid: A large-scale dataset for wildlife detection with drones." Applied Sciences 13.18 (2023): 10397.
Google Earth: Satellite and aerial imagery obtained through screenshots from Google Earth. Imagery © Google and respective third-party imagery providers.

Google Earth Imagery Attribution

Some images included in BMRS were collected as screenshots from Google Earth.

The original imagery metadata, copyright attribution, and source information were not preserved during dataset preparation. Consequently, it is not possible to reliably identify the specific imagery providers, acquisition dates, or attribution information for individual samples.

The underlying imagery remains subject to the applicable copyright and licensing terms of Google and its respective third-party imagery providers.

For further information, please refer to:

Citation

@article{202606.1484,
    doi = {10.20944/preprints202606.1484.v1},
    url = {https://doi.org/10.20944/preprints202606.1484.v1},
    year = 2026,
    month = {June},
    publisher = {Preprints},
    author = {Nikita Firsov and Olga Terekhova and Nikita Odinets and Alexey Fedotov and Artem Muzyka and Anna Ukhanaeva and Anastasia Sarycheva and Sergei Gladilin and Dmitry Sidorchuk},
    title = {BMRS: Bongard–Maximov Problems for Remote Sensing},
    journal = {Preprints}
}

License

This dataset is released under license: cc-by-4.0

Third-party imagery included in BMRS may be subject to separate copyright and licensing restrictions. Google Earth imagery remains subject to the applicable terms of Google and its respective third-party imagery providers and is not relicensed under CC BY 4.0 by the BMRS authors.

Downloads last month: 83