Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
image
imagewidth (px)
256
1.02k
label
class label
1.09k classes
0004cca639e2fce1c
100a30ac0e49eb8c5
200efba79f8c7abed
3012eb827f27e53ea
401f6f24491c72296
501ffa48e1cfdcc60
60222907bd4fd7a2e
7023b377503332cb7
8023d5eac5d202fdf
9024027ac85482ac2
1002856635798a07ac
1102b4109b62d570e7
1202fab9159658eb74
13033176ed1766ded2
1403c3c5809e2c6863
1503d6a60385313603
1603dc31cd6be16ef8
170427aee1898e420a
18043037d1af166806
19046ae6a52404cc5e
20048491c0a489122b
2104a249c5c8a2131b
2204d9d687759607f5
2304e042276139c9f8
24053b0ac9ec57c923
2505cac223691b96ee
2605e8459da9995d06
27063af19a91a1f52c
28069fe121df98c227
2906d4564532ccd587
30075fb2d9e6505b40
310773d14fbe0e08b9
320777e6105ec30a72
33077d9679d389cced
340796640eae23c7ff
3507aca1548e875d12
3607b2786e05f43507
3707bd104f0d06049a
3807eb0b750e37c2f6
3908a3c6f5a460a4e3
4008ce74f01eb3b2ff
410925c3e80fe462fc
4209bd147fb28cedc2
4309df66a2344f9b21
440a6b9e95651c3947
450a90dc5a016a41ae
460ab7abe9fb830a66
470ac0a49b18ac1a20
480b18e258383dd6b2
490b3d9187387b4c3b
500b6300af87d61cad
510c14d11d0c3df1dc
520c180b18157aafdb
530c2efce553c8aa94
540c6657b6a4a459da
550ca9eacb31d08ecd
560d850cac5654fa52
570d8826898481dafe
580ddca8c615eb5b1d
590eadf7983000f211
600ed32c8043f797cd
610eda9ea7998d04e2
620edf53b4926d7b42
630f0a9ea804ff565a
640f3d3f2e0a4072ff
650f3e04b834d68d40
660f750b95944560c4
670f792ed7c49ad0bf
680fff3cd3482f4158
69100a495d3ec12c3d
7010238a5d1d9921c0
711042d8b21a73d6bd
721058fc89ee5d63b7
7310907615ed7fb787
7411280f7c8632950e
75114ccef0bed46ac8
7611cccad00e379cb3
7711e62aa8f77ec86b
7812043a891f5cad29
79120f39fc840a1cbc
801215cffd118a58be
8112f777cd99667e4c
8212fce11aabb20002
83137b90638a0454a0
841399531bf081ad13
851438e60b6b30af1f
86145400c5e44169d9
8714545efacc67eb5f
88145bfe5d5760150f
891469956a6d8c3fef
9014b2b6dde865db87
91150a255a5f9332d5
92153ff16253006996
9315b0c54826c4c964
9415bf4f475ef879bb
9515cebf441fd35db5
9615fcdb08c745787c
9716158555079a7e0f
98161ca5e3f39f1b8b
991671edd36854ac28
End of preview. Expand in Data Studio

OpenUSD Explorer transparent previews

Transparent PNG derivatives of original OpenUSD asset thumbnails, prepared for OpenUSD Explorer. The original viewpoint and framing are preserved. No 3D model files are hosted here. Native-alpha previews remain links to their source datasets.

Preparation uses the local Python package rembg 2.0.86 with BiRefNet-general and edge-color decontamination on NVIDIA T4. It does not use a hosted removal API. The model's SHA-256 and batch settings are recorded in the preparation reports.

All cutouts were screened on contact sheets, with larger source/cutout comparisons for flagged and selected detailed objects. This is thumbnail-level QA, not a guarantee of physically correct transparency: glass can retain shading baked into the original image, and thin geometry is limited by source resolution.

Attribution and licenses

Each manifest entry identifies its source dataset, creator through the source URL, exact source revision, original image URL, declared asset license and image hash. Attribution and original license terms remain applicable to each image. An absent license field means the upstream metadata did not specify one; it does not mean public domain. Native linked images are not redistributed in this repo. No single application-code license is applied to this mixed-source collection.

The transformation is background removal, with any per-image correction described in its review record. Source asset owners retain their respective rights.

Downloads last month
-

Space using fffiloni/openusd-explorer-previews 1