VIEW2SPACE
Collection
Data & Checkpoint of ECCV26 paper: VIEW2SPACE • 4 items • Updated • 1
Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: TypeError
Message: Couldn't cast array of type struct<images/img_012633.png: list<item: null>, images/img_013066.png: list<item: null>, images/img_019083.png: list<item: null>, images/img_014930.png: list<item: null>, images/img_008989.png: list<item: null>, images/img_012244.png: list<item: null>, images/img_014812.png: list<item: null>, images/img_019197.png: list<item: null>, images/img_009951.png: list<item: null>, images/img_012311.png: list<item: null>, images/img_013301.png: list<item: null>, images/img_013170.png: list<item: null>, images/img_013540.png: list<item: null>, images/img_019079.png: list<item: null>, images/img_012773.png: list<item: null>, images/img_016846.png: list<item: null>, images/img_019062.png: list<item: null>, images/img_013966.png: list<item: null>, images/img_014117.png: list<item: null>, images/img_013216.png: list<item: null>, images/img_019283.png: list<item: null>, images/img_009087.png: list<item: null>, images/img_013221.png: list<item: null>, images/img_010781.png: list<item: null>, images/img_018681.png: list<item: null>, images/img_012225.png: list<item: null>, images/img_010866.png: list<item: null>, images/img_019214.png: list<item: null>, images/img_013185.png: list<item: null>, images/img_013525.png: list<item: null>, images/img_011610.png: list<item: null>, images/img_012178.png: list<item: null>, images/img_011609.png: list<item: null>, images/img_010948.png: list<item: null>, images/img_019381.png: list<item: null>, images/img_014968.png: list<item: null>, images/img_019296
...
5.png: list<item: null>, images/img_012443.png: list<item: null>, images/img_010985.png: list<item: null>, images/img_012713.png: list<item: null>, images/img_013678.png: list<item: null>, images/img_009542.png: list<item: null>, images/img_014137.png: list<item: null>, images/img_013152.png: list<item: null>, images/img_010986.png: list<item: null>, images/img_018044.png: list<item: null>, images/img_011607.png: list<item: null>, images/img_009949.png: list<item: null>, images/img_011477.png: list<item: null>, images/img_011789.png: list<item: null>, images/img_012598.png: list<item: null>, images/img_009970.png: list<item: null>, images/img_012950.png: list<item: null>, images/img_019367.png: list<item: null>, images/img_014030.png: list<item: null>, images/img_013153.png: list<item: null>, images/img_012324.png: list<item: null>, images/img_014253.png: list<item: null>, images/img_012309.png: list<item: null>, images/img_012359.png: list<item: null>, images/img_017842.png: list<item: null>, images/img_010675.png: list<item: null>, images/img_014256.png: list<item: null>, images/img_009698.png: list<item: null>, images/img_009744.png: list<item: null>, images/img_010952.png: list<item: null>, images/img_018734.png: list<item: null>, images/img_009347.png: list<item: null>, images/img_019738.png: list<item: null>, images/img_014998.png: list<item: null>, images/img_014292.png: list<item: null>, images/img_012827.png: list<item: null>, images/img_009262.png: list<item: null>> to null
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1816, in _prepare_split_single
for key, table in generator:
^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 613, in wrapped
for item in generator(*args, **kwargs):
~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2369, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2303, in cast_table_to_schema
cast_array_to_feature(
~~~~~~~~~~~~~~~~~~~~~^
table[name] if name in table_column_names else pa.array([None] * len(table), type=schema.field(name).type),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
feature,
^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1852, in wrapper
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2059, in cast_array_to_feature
_c(array.field(name) if name in array_fields else null_array, subfeature)
~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1854, in wrapper
return func(array, *args, **kwargs)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2143, in cast_array_to_feature
return array_cast(
array,
...<2 lines>...
allow_decimal_to_str=allow_decimal_to_str,
)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1854, in wrapper
return func(array, *args, **kwargs)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2007, in array_cast
raise TypeError(f"Couldn't cast array of type {_short_str(array.type)} to {_short_str(pa_type)}")
TypeError: Couldn't cast array of type struct<images/img_012633.png: list<item: null>, images/img_013066.png: list<item: null>, images/img_019083.png: list<item: null>, images/img_014930.png: list<item: null>, images/img_008989.png: list<item: null>, images/img_012244.png: list<item: null>, images/img_014812.png: list<item: null>, images/img_019197.png: list<item: null>, images/img_009951.png: list<item: null>, images/img_012311.png: list<item: null>, images/img_013301.png: list<item: null>, images/img_013170.png: list<item: null>, images/img_013540.png: list<item: null>, images/img_019079.png: list<item: null>, images/img_012773.png: list<item: null>, images/img_016846.png: list<item: null>, images/img_019062.png: list<item: null>, images/img_013966.png: list<item: null>, images/img_014117.png: list<item: null>, images/img_013216.png: list<item: null>, images/img_019283.png: list<item: null>, images/img_009087.png: list<item: null>, images/img_013221.png: list<item: null>, images/img_010781.png: list<item: null>, images/img_018681.png: list<item: null>, images/img_012225.png: list<item: null>, images/img_010866.png: list<item: null>, images/img_019214.png: list<item: null>, images/img_013185.png: list<item: null>, images/img_013525.png: list<item: null>, images/img_011610.png: list<item: null>, images/img_012178.png: list<item: null>, images/img_011609.png: list<item: null>, images/img_010948.png: list<item: null>, images/img_019381.png: list<item: null>, images/img_014968.png: list<item: null>, images/img_019296
...
5.png: list<item: null>, images/img_012443.png: list<item: null>, images/img_010985.png: list<item: null>, images/img_012713.png: list<item: null>, images/img_013678.png: list<item: null>, images/img_009542.png: list<item: null>, images/img_014137.png: list<item: null>, images/img_013152.png: list<item: null>, images/img_010986.png: list<item: null>, images/img_018044.png: list<item: null>, images/img_011607.png: list<item: null>, images/img_009949.png: list<item: null>, images/img_011477.png: list<item: null>, images/img_011789.png: list<item: null>, images/img_012598.png: list<item: null>, images/img_009970.png: list<item: null>, images/img_012950.png: list<item: null>, images/img_019367.png: list<item: null>, images/img_014030.png: list<item: null>, images/img_013153.png: list<item: null>, images/img_012324.png: list<item: null>, images/img_014253.png: list<item: null>, images/img_012309.png: list<item: null>, images/img_012359.png: list<item: null>, images/img_017842.png: list<item: null>, images/img_010675.png: list<item: null>, images/img_014256.png: list<item: null>, images/img_009698.png: list<item: null>, images/img_009744.png: list<item: null>, images/img_010952.png: list<item: null>, images/img_018734.png: list<item: null>, images/img_009347.png: list<item: null>, images/img_019738.png: list<item: null>, images/img_014998.png: list<item: null>, images/img_014292.png: list<item: null>, images/img_012827.png: list<item: null>, images/img_009262.png: list<item: null>> to null
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
~~~~~~~~~~~~~~~~~~~~~~~~~^
builder, max_dataset_size_bytes=max_dataset_size_bytes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1683, in _prepare_split
for job_id, done, content in self._prepare_split_single(
~~~~~~~~~~~~~~~~~~~~~~~~~~^
gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
):
^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1869, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
q_idx string | q_type string | question string | options unknown | question_prompt string | answer int64 | image_paths list | supporting dict |
|---|---|---|---|---|---|---|---|
count_000001 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many white dog is present in this scene? | {} | 1 | [
"images/img_010168.png",
"images/img_010170.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the litter box (instance_1) is at <bbox_1> [468.0, 717.0, 484.0, 747.0] </bbox_1>; the bin (instance_2) is at <bbox_1> [624.0, 687.0, 642.0, 743.0] </bbox_1>; the litter bo... | |
count_000002 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many vise are present in this scene? | {} | 4 | [
"images/img_017457.png",
"images/img_017461.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the storage rack with four shelves (instance_1) is at <bbox_1> [769.0, 567.0, 884.0, 747.0] </bbox_1>; the storage rack with four shelves (instance_2) is at <bbox_1> [605.0... | |
count_000003 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many drone with angular white body and black propellers is present in this scene? | {} | 1 | [
"images/img_011956.png",
"images/img_011958.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the drone with angular white body and black propellers (instance_1) is at <bbox_1> [581.0, 450.0, 635.0, 505.0] </bbox_1>; the drone with rectangular white and black body (... | |
count_000004 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many human with yellow overalls are present in this scene? | {} | 3 | [
"images/img_016535.png",
"images/img_016533.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the human with plaid shirt (instance_1) is at <bbox_1> [755.0, 420.0, 833.0, 572.0] </bbox_1>; the human with orange jacket (instance_2) is at <bbox_1> [560.0, 510.0, 599.0... | |
count_000005 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many tea set are present in this scene? | {} | 2 | [
"images/img_009842.png",
"images/img_009838.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the blue potted plant (instance_1) is at <bbox_1> [690.0, 353.0, 739.0, 438.0] </bbox_1>; the tea set (instance_2) is at <bbox_1> [637.0, 370.0, 649.0, 378.0] </bbox_1>; th... | |
count_000006 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many blue shipping container are present in this scene? | {} | 2 | [
"images/img_015299.png",
"images/img_015294.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the truck with white and black cab (instance_1) is at <bbox_1> [611.0, 595.0, 770.0, 795.0] </bbox_1>; the white shipping container (instance_2) is at <bbox_1> [500.0, 575.... | |
count_000007 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many patrol boat with white foredeck are present in this scene? | {} | 2 | [
"images/img_012556.png",
"images/img_012550.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the orange motorboat (instance_1) is at <bbox_1> [534.0, 543.0, 576.0, 582.0] </bbox_1>; the red motorboat (instance_2) is at <bbox_1> [665.0, 463.0, 706.0, 492.0] </bbox_1... | |
count_000008 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many dog are present in this scene? | {} | 6 | [
"images/img_010296.png",
"images/img_010295.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the dog (instance_1) is at <bbox_1> [487.0, 255.0, 501.0, 275.0] </bbox_1>; the dog (instance_2) is at <bbox_1> [704.0, 618.0, 735.0, 677.0] </bbox_1>; the doghouse (instan... | |
count_000009 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many truck are present in this scene? | {} | 6 | [
"images/img_015299.png",
"images/img_015300.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the truck (instance_1) is at <bbox_1> [611.0, 595.0, 770.0, 795.0] </bbox_1>; the white shipping container (instance_2) is at <bbox_1> [500.0, 575.0, 614.0, 713.0] </bbox_1... | |
count_000010 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many table with white tabletop are present in this scene? | {} | 4 | [
"images/img_018030.png",
"images/img_018034.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the person seated in white (instance_1) is at <bbox_1> [635.0, 475.0, 665.0, 607.0] </bbox_1>; the Evian water bottle (instance_2) is at <bbox_1> [634.0, 728.0, 640.0, 748.... | |
count_000011 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many toy are present in this scene? | {} | 24 | [
"images/img_013179.png",
"images/img_013171.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the toy (instance_1) is at <bbox_1> [521.0, 303.0, 598.0, 378.0] </bbox_1>; the toy (instance_2) is at <bbox_1> [941.0, 340.0, 966.0, 385.0] </bbox_1>; the toy (instance_3)... | |
count_000012 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many water tower are present in this scene? | {} | 5 | [
"images/img_018436.png",
"images/img_018433.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the water tower (instance_1) is at <bbox_1> [242.0, 185.0, 341.0, 430.0] </bbox_1>; the gray spherical storage tank (instance_2) is at <bbox_1> [424.0, 615.0, 490.0, 777.0]... | |
count_000013 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many modern two-story house with a flat roof, white walls, and attached garage is present in this scene? | {} | 0 | [
"images/img_014892.png",
"images/img_014894.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the house with flat roof (instance_1) is at <bbox_1> [5.0, 340.0, 152.0, 643.0] </bbox_1>; the yellow coupe (instance_2) is at <bbox_1> [199.0, 740.0, 335.0, 820.0] </bbox_... | |
count_000014 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many stone pot are present in this scene? | {} | 2 | [
"images/img_016006.png",
"images/img_015998.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the canopy stand with blue and yellow strip canopy (instance_1) is at <bbox_1> [599.0, 645.0, 800.0, 853.0] </bbox_1>; the tiger shape pot (instance_2) is at <bbox_1> [754.... | |
count_000015 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many pistol are present in this scene? | {} | 2 | [
"images/img_013126.png",
"images/img_013125.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the target (instance_1) is at <bbox_1> [735.0, 530.0, 861.0, 752.0] </bbox_1>; the barrel (instance_2) is at <bbox_1> [339.0, 645.0, 406.0, 787.0] </bbox_1>; the submachine... | |
count_000016 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many fighter jet are present in this scene? | {} | 8 | [
"images/img_018474.png",
"images/img_018467.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the fighter jet (instance_1) is at <bbox_1> [448.0, 695.0, 598.0, 910.0] </bbox_1>; the green and black striped hot air balloon (instance_2) is at <bbox_1> [297.0, 333.0, 4... | |
count_000017 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many rocking chair are present in this scene? | {} | 2 | [
"images/img_013993.png",
"images/img_013990.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the wooden cabinet (instance_1) is at <bbox_1> [374.0, 503.0, 434.0, 662.0] </bbox_1>; the table with brown tabletop (instance_2) is at <bbox_1> [398.0, 357.0, 466.0, 467.0... | |
count_000018 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many car are present in this scene? | {} | 16 | [
"images/img_014892.png",
"images/img_014886.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the car (instance_1) is at <bbox_1> [307.0, 353.0, 390.0, 398.0] </bbox_1>; the house with flat roof (instance_2) is at <bbox_1> [5.0, 340.0, 152.0, 643.0] </bbox_1>; the c... | |
count_000019 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many airplane with white and green livery is present in this scene? | {} | 1 | [
"images/img_011885.png",
"images/img_011887.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the stylized F22 fighter jet with white and red color (instance_1) is at <bbox_1> [502.0, 343.0, 571.0, 392.0] </bbox_1>; the airplane with yellow livery (instance_2) is at... | |
count_000020 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many table with white tabletop are present in this scene? | {} | 4 | [
"images/img_017840.png",
"images/img_017839.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the bar counter (instance_1) is at <bbox_1> [616.0, 228.0, 999.0, 602.0] </bbox_1>; the person (instance_2) is at <bbox_1> [376.0, 702.0, 429.0, 888.0] </bbox_1>\nIn view 2... | |
count_000021 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many fighter jet are present in this scene? | {} | 8 | [
"images/img_018539.png",
"images/img_018540.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the fighter jet (instance_1) is at <bbox_1> [211.0, 365.0, 334.0, 468.0] </bbox_1>; the red and black striped hot air balloon (instance_2) is at <bbox_1> [0.0, 377.0, 177.0... | |
count_000022 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many drone are present in this scene? | {} | 10 | [
"images/img_012016.png",
"images/img_012017.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the drone (instance_1) is at <bbox_1> [375.0, 758.0, 464.0, 857.0] </bbox_1>; the drone (instance_2) is at <bbox_1> [235.0, 508.0, 307.0, 573.0] </bbox_1>; the drone (insta... | |
count_000023 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many black cylindrical storage tank are present in this scene? | {} | 4 | [
"images/img_018207.png",
"images/img_018212.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the water tower with gray spherical top (instance_1) is at <bbox_1> [443.0, 257.0, 476.0, 363.0] </bbox_1>; the black cylindrical storage tank (instance_2) is at <bbox_1> [... | |
count_000024 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many white supercar are present in this scene? | {} | 2 | [
"images/img_015166.png",
"images/img_015161.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the house (instance_1) is at <bbox_1> [19.0, 253.0, 349.0, 580.0] </bbox_1>; the white supercar (instance_2) is at <bbox_1> [511.0, 852.0, 674.0, 965.0] </bbox_1>; the whit... | |
count_000025 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many camera are present in this scene? | {} | 3 | [
"images/img_013875.png",
"images/img_013869.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the wooden cabinet (instance_1) is at <bbox_1> [441.0, 418.0, 509.0, 592.0] </bbox_1>; the wooden cabinet (instance_2) is at <bbox_1> [664.0, 307.0, 723.0, 440.0] </bbox_1>... | |
count_000026 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many supercar are present in this scene? | {} | 5 | [
"images/img_015114.png",
"images/img_015111.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the two-story modern house with a flat roof, white walls, and grass strips along the driveway (instance_1) is at <bbox_1> [629.0, 35.0, 1000.0, 395.0] </bbox_1>; the superc... | |
count_000027 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many green shipping container are present in this scene? | {} | 2 | [
"images/img_015319.png",
"images/img_015315.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the white truck with flatbed (instance_1) is at <bbox_1> [877.0, 178.0, 917.0, 217.0] </bbox_1>; the green shipping container (instance_2) is at <bbox_1> [527.0, 598.0, 580... | |
count_000028 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many palm tree is present in this scene? | {} | 0 | [
"images/img_016838.png",
"images/img_016832.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the lifeguard tower (instance_1) is at <bbox_1> [675.0, 377.0, 999.0, 997.0] </bbox_1>; the beach ball (instance_2) is at <bbox_1> [511.0, 352.0, 544.0, 395.0] </bbox_1>; t... | |
count_000029 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many stylized F22 fighter jet with grey and cyan color are present in this scene? | {} | 2 | [
"images/img_018519.png",
"images/img_018522.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the stylized F22 fighter jet with white and red color (instance_1) is at <bbox_1> [196.0, 403.0, 334.0, 487.0] </bbox_1>; the green and yellow striped hot air balloon (inst... | |
count_000030 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many drone with angular white body and gray propellers are present in this scene? | {} | 2 | [
"images/img_011980.png",
"images/img_011976.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the drone with rectangular white and black body (instance_1) is at <bbox_1> [479.0, 772.0, 574.0, 880.0] </bbox_1>; the drone with angular black body (instance_2) is at <bb... | |
count_000031 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many fighter jet are present in this scene? | {} | 8 | [
"images/img_018585.png",
"images/img_018591.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the fighter jet (instance_1) is at <bbox_1> [641.0, 348.0, 766.0, 403.0] </bbox_1>; the white and green striped hot air balloon (instance_2) is at <bbox_1> [456.0, 202.0, 5... | |
count_000032 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many stylized F22 fighter jet with white and red color are present in this scene? | {} | 3 | [
"images/img_018779.png",
"images/img_018772.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the F-16 fighter jet (instance_1) is at <bbox_1> [169.0, 473.0, 329.0, 585.0] </bbox_1>; the green and yellow striped hot air balloon (instance_2) is at <bbox_1> [816.0, 43... | |
count_000033 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many red barrel are present in this scene? | {} | 4 | [
"images/img_013060.png",
"images/img_013055.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the target (instance_1) is at <bbox_1> [491.0, 203.0, 562.0, 370.0] </bbox_1>; the red barrel (instance_2) is at <bbox_1> [656.0, 387.0, 710.0, 500.0] </bbox_1>; the ammuni... | |
count_000034 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many bench are present in this scene? | {} | 4 | [
"images/img_015685.png",
"images/img_015680.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the swing (instance_1) is at <bbox_1> [739.0, 42.0, 858.0, 53.0] </bbox_1>; the basketball (instance_2) is at <bbox_1> [385.0, 392.0, 404.0, 417.0] </bbox_1>; the beach bal... | |
count_000035 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many red motorboat are present in this scene? | {} | 6 | [
"images/img_012706.png",
"images/img_012710.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the patrol boat with white foredeck (instance_1) is at <bbox_1> [470.0, 435.0, 510.0, 533.0] </bbox_1>; the inflatable boat (instance_2) is at <bbox_1> [289.0, 398.0, 324.0... | |
count_000036 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many desert camouflage helicopter are present in this scene? | {} | 2 | [
"images/img_019153.png",
"images/img_019159.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the beige sand helicopter (instance_1) is at <bbox_1> [371.0, 638.0, 615.0, 803.0] </bbox_1>; the red and black striped hot air balloon (instance_2) is at <bbox_1> [0.0, 35... | |
count_000037 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many tea set is present in this scene? | {} | 1 | [
"images/img_009788.png",
"images/img_009786.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the umbrella (instance_1) is at <bbox_1> [650.0, 482.0, 704.0, 538.0] </bbox_1>; the marble pedestal with round legs (instance_2) is at <bbox_1> [613.0, 305.0, 681.0, 380.0... | |
count_000038 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many rigid black metal bench with armrests are present in this scene? | {} | 2 | [
"images/img_015680.png",
"images/img_015679.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the play structure (instance_1) is at <bbox_1> [647.0, 592.0, 1000.0, 1000.0] </bbox_1>; the swing with two seats (instance_2) is at <bbox_1> [226.0, 65.0, 386.0, 252.0] </... | |
count_000039 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many clock are present in this scene? | {} | 4 | [
"images/img_014057.png",
"images/img_014060.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the table with brown tabletop (instance_1) is at <bbox_1> [469.0, 298.0, 529.0, 395.0] </bbox_1>; the cabinet (instance_2) is at <bbox_1> [741.0, 392.0, 807.0, 533.0] </bbo... | |
count_000040 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many lavender plant with white pot are present in this scene? | {} | 8 | [
"images/img_010994.png",
"images/img_010996.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the lavender plant with green pot (instance_1) is at <bbox_1> [266.0, 542.0, 310.0, 615.0] </bbox_1>; the sunflower plant with white pot (instance_2) is at <bbox_1> [504.0,... | |
count_000041 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many human with red shirt is present in this scene? | {} | 1 | [
"images/img_016640.png",
"images/img_016633.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the human with red shirt (instance_1) is at <bbox_1> [945.0, 667.0, 1000.0, 778.0] </bbox_1>; the air hockey table with blue surface (instance_2) is at <bbox_1> [897.0, 187... | |
count_000042 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many table with black tabletop are present in this scene? | {} | 2 | [
"images/img_014056.png",
"images/img_014063.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the table with brown tabletop (instance_1) is at <bbox_1> [279.0, 325.0, 386.0, 437.0] </bbox_1>; the cabinet (instance_2) is at <bbox_1> [659.0, 310.0, 713.0, 438.0] </bbo... | |
count_000043 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many storage tank are present in this scene? | {} | 8 | [
"images/img_018436.png",
"images/img_018439.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the water tower with blue spherical top (instance_1) is at <bbox_1> [242.0, 185.0, 341.0, 430.0] </bbox_1>; the storage tank (instance_2) is at <bbox_1> [424.0, 615.0, 490.... | |
count_000044 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many human with orange dress are present in this scene? | {} | 3 | [
"images/img_016403.png",
"images/img_016400.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the human with black dress (instance_1) is at <bbox_1> [297.0, 602.0, 339.0, 735.0] </bbox_1>; the human with beige jacket and blue jeans (instance_2) is at <bbox_1> [806.0... | |
count_000045 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many person lying on workout mat is present in this scene? | {} | 1 | [
"images/img_018838.png",
"images/img_018841.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the kettlebell with blue marker (instance_1) is at <bbox_1> [130.0, 450.0, 146.0, 490.0] </bbox_1>; the person holding rolled mat (instance_2) is at <bbox_1> [914.0, 322.0,... | |
count_000046 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many black drum set is present in this scene? | {} | 0 | [
"images/img_012332.png",
"images/img_012340.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the white table (instance_1) is at <bbox_1> [349.0, 493.0, 677.0, 877.0] </bbox_1>; the guitar with orange body (instance_2) is at <bbox_1> [448.0, 548.0, 505.0, 692.0] </b... | |
count_000047 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many kettlebell with green marker are present in this scene? | {} | 3 | [
"images/img_018838.png",
"images/img_018840.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the kettlebell with blue marker (instance_1) is at <bbox_1> [130.0, 450.0, 146.0, 490.0] </bbox_1>; the person holding rolled mat (instance_2) is at <bbox_1> [914.0, 322.0,... | |
count_000048 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many bottle are present in this scene? | {} | 6 | [
"images/img_017921.png",
"images/img_017917.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the bottle (instance_1) is at <bbox_1> [466.0, 437.0, 480.0, 490.0] </bbox_1>; the neon sign (instance_2) is at <bbox_1> [780.0, 815.0, 859.0, 1000.0] </bbox_1>; the person... | |
count_000049 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many hot air balloon are present in this scene? | {} | 4 | [
"images/img_019224.png",
"images/img_019226.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the green camouflage helicopter (instance_1) is at <bbox_1> [193.0, 405.0, 302.0, 463.0] </bbox_1>; the hot air balloon (instance_2) is at <bbox_1> [0.0, 285.0, 121.0, 503.... | |
count_000050 | count | All views are captured from the same static scene arrangement, but from different positions and angles.
How many cauldron is present in this scene? | {} | 1 | [
"images/img_011072.png",
"images/img_011077.png"
] | {
"draw_boxes": null,
"chain_of_thought": "<think> Let me observe instances across views (same instance_id = same physical object).\nIn view 1: the tombstone (instance_1) is at <bbox_1> [323.0, 592.0, 351.0, 672.0] </bbox_1>; the guillotine (instance_2) is at <bbox_1> [446.0, 713.0, 544.0, 935.0] </bbox_1>; the car... |
End of preview.
VIEW2SPACE Training
This repository contains the public VIEW2SPACE training release for multi-view vision-language spatial reasoning.
Associated paper:
- VIEW2SPACE: Studying Multi-View Visual Reasoning from Sparse Observations - ECCV 2026 🚀
- arXiv: 2603.16506
The training release contains:
604,779training examples22,205public images- three public question families:
count,detect, andmcq
Unlike the evaluation release, this training package keeps supporting.chain_of_thought so it can be used for supervision.
Related VIEW2SPACE Releases
- Testing release: Pokerme/view2space-v1
- 4B model checkpoint: Pokerme/view2space_4b
For the official project, updates, and citation information, see:
Directory Structure
view2space-train-release/
README.md
overall.jsonl
images/
img_000001.png
img_000002.png
...
Each line in overall.jsonl is one training example.
Data Format
Each JSONL record contains the following public fields:
q_idx: anonymized question id such ascount_000001q_type: public question family, one ofcount,detect, ormcqquestion: question textoptions: multiple-choice options for MCQ questions; empty for non-MCQ questionsquestion_prompt: extra prompt text when applicableanswer: ground-truth answerimage_paths: relative paths to the images used by the questionsupporting.draw_boxes: optional input boxes associated with one or more imagessupporting.chain_of_thought: released reasoning trace for training
Notes:
image_pathsare relative to the dataset root.- Example image path:
images/img_000123.png - original internal scene names, image names, and fine-grained question types are anonymized in the public release
Example Record
{
"q_idx": "mcq_000001",
"q_type": "mcq",
"question": "Based on these four images (image 1, 2, 3, and 4) showing the ball from different viewpoints (front, left, back, and right), with each camera aligned with walls and partially capturing the surroundings: \nFrom the viewpoint presented in image 4, what is to the left of ball?",
"options": {
"A": "tv",
"B": "fountain",
"C": "sofa",
"D": "table"
},
"question_prompt": "",
"answer": "B",
"image_paths": [
"images/img_000001.png",
"images/img_000002.png",
"images/img_000003.png",
"images/img_000004.png"
],
"supporting": {
"draw_boxes": null,
"chain_of_thought": "<think> ... </think>\n <answer> B </answer>"
}
}
Basic Usage
import json
from pathlib import Path
dataset_root = Path("/path/to/view2space-train-release")
jsonl_path = dataset_root / "overall.jsonl"
with jsonl_path.open("r", encoding="utf-8") as f:
first = json.loads(next(f))
image_files = [dataset_root / rel_path for rel_path in first["image_paths"]]
reasoning = first["supporting"]["chain_of_thought"]
print(first["q_idx"])
print(first["q_type"])
print(image_files)
print(reasoning[:200])
Citation
License and Citation
This dataset is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0). Attribution is required under the license. For academic and research use, citation in the resulting paper, report, model card, dataset card, or public documentation is expected. Please cite this dataset whenever it is used for training, evaluation, benchmarking, data analysis, or as part of a larger dataset mixture.
@article{ke2026view2space,
title={VIEW2SPACE: Studying Multi-View Visual Reasoning from Sparse Observations},
author={Ke, Fucai and Cai, Zhixi and Li, Boying and Chen, Long and Lin, Beibei and Wang, Weiqing and Haghighi, Pari Delir and Haffari, Gholamreza and Rezatofighi, Hamid},
journal={arXiv preprint arXiv:2603.16506},
year={2026}
}
- Downloads last month
- 49