stemjara commited on
Commit
207342b
·
verified ·
1 Parent(s): d2fed4e

Card: Sundial Lite renders, second round, split notes

Browse files
Files changed (1) hide show
  1. README.md +174 -126
README.md CHANGED
@@ -1,126 +1,174 @@
1
- ---
2
- license: other
3
- license_name: mixed
4
- license_details: Annotations CC BY 4.0; images are Minecraft screenshots under Mojang's usage guidelines
5
- pretty_name: Minecraft vanilla frames with depth, camera and traced sun shadows
6
- task_categories:
7
- - image-to-image
8
- - depth-estimation
9
- tags:
10
- - minecraft
11
- - webdataset
12
- - depth
13
- - shadows
14
- size_categories:
15
- - 10K<n<100K
16
- configs:
17
- - config_name: preview
18
- default: true
19
- data_files:
20
- - split: train
21
- path: preview/train-*.tar
22
- - split: test
23
- path: preview/test-*.tar
24
- ---
25
-
26
- # Minecraft vanilla frames with depth, camera and traced sun shadows
27
-
28
- 16 364 frames of Minecraft Java Edition 1.21.1 rendered **without shaders** at 1280×720, each with
29
- its depth buffer, its camera, the block under every pixel and the sun shadow traced in the saved
30
- world. They were captured with a Fabric mod that teleports the player through a plan of poses with
31
- the world frozen (no weather, no entities, fixed time of day).
32
-
33
- Capture tool (the mod, its configuration, the shader options and the contribution validator):
34
- **<https://github.com/stevenijm777/nrmc-capture>**
35
-
36
- *Resumen en español: 16 364 fotogramas de Minecraft 1.21.1 sin shaders, con profundidad, cámara,
37
- bloque por píxel y la sombra del sol trazada en el mundo guardado. Solo el pase vanilla.*
38
-
39
- ## What is not here
40
-
41
- This is the input half of a paired dataset (vanilla frame → the same frame rendered with a shader
42
- pack). The shader renders are **not included**: their licence does not clearly allow
43
- redistributing them. With the poses and seeds below and the
44
- [capture tool](https://github.com/stevenijm777/nrmc-capture), the same frames can be rendered again
45
- with any shader pack.
46
-
47
- ## Splits
48
-
49
- | Split | Scenes | World |
50
- |---|---|---|
51
- | train | 13 039 | `minesr_atlas` |
52
- | test | 2 625 | `minesr_atlas`, in cells of the map not used for training |
53
- | test | 700 | `minesr_test2` (swamp), a different world |
54
-
55
- | Biome | Scenes |
56
- |---|---|
57
- | forest | 3 990 |
58
- | swamp | 3 430 |
59
- | taiga | 3 346 |
60
- | plains | 2 800 |
61
- | desert | 1 400 |
62
- | snowy_plains | 1 398 |
63
-
64
- World seeds: `minesr_atlas` 5277617577473417202, `minesr_test2` 7308878374753708357.
65
-
66
- Every position is captured at several times of day (`time_of_day`, in game ticks). 224 scenes
67
- (32 positions) were removed by hand because an animal or a dropped item was in view.
68
-
69
- ## Files
70
-
71
- WebDataset shards, up to 500 scenes and about 1 GB each: `data/<split>/<world>-<biome>-<nnn>.tar`.
72
- `manifest.jsonl` has one line per scene (pose, biome, time, split and the shard that holds it).
73
-
74
- The dataset viewer shows `preview/`: a 640×360 JPEG thumbnail and the manifest line of every
75
- scene. It is only for browsing; the data is in `data/`.
76
-
77
- Inside a shard, per scene:
78
-
79
- | File | Content |
80
- |---|---|
81
- | `<scene_id>.vanilla.png` | the frame, RGB, 1280×720 |
82
- | `<scene_id>.depth.npz` | `depth` float32 (720, 1280): OpenGL window-space depth in [0, 1], 1 = far plane or sky, origin top left |
83
- | `<scene_id>.json` | pose, biome, weather, `time_of_day`, and `camera`: `position`, `view_matrix` and `projection_matrix` (4×4, column major) |
84
- | `<scene_id>.blocks.npz` | `ids` uint16 (720, 1280): index into `names`, the block under each pixel (`nrmc:sky` for the sky) |
85
- | `<scene_id>.shadow.npz` | sun shadow of each pixel, traced through the blocks of the saved world (see below) |
86
-
87
- Linearise the depth with `camera.projection_matrix`; with the view matrix this gives the world
88
- position of every pixel.
89
-
90
- ### `shadow.npz`
91
-
92
- One ray per pixel from its surface towards the sun, through the block grid of the saved world.
93
- This is information the renderer has and the image alone does not: the occluder can be outside
94
- the frame.
95
-
96
- | Array | Content |
97
- |---|---|
98
- | `bits` uint8 | flags of what the ray crossed: 1 solid block, 2 leaves, 4 plant; 8 the first occluder projects inside the frame; 16 it is also visible in the frame |
99
- | `first` float16 | distance in blocks to the first occluder, `inf` if none |
100
- | `leaves` uint8 | number of leaf cells crossed (capped at 255) |
101
- | `screen_test` uint8 | internal flag of the tracer |
102
-
103
- ## Loading
104
-
105
- Download the shards (`hf download stemjara/minecraft-vanilla-frames --repo-type dataset
106
- --include "data/*"`) and read them with `webdataset` or `tarfile`:
107
-
108
- ```python
109
- import io, json, numpy as np, webdataset as wds
110
-
111
- ds = wds.WebDataset("data/train/minesr_atlas-forest-000.tar")
112
- for s in ds:
113
- meta = json.loads(s["json"])
114
- depth = np.load(io.BytesIO(s["depth.npz"]))["depth"]
115
- shadow = np.load(io.BytesIO(s["shadow.npz"]))
116
- png_bytes = s["vanilla.png"]
117
- break
118
- ```
119
-
120
- ## Licence
121
-
122
- - The images are screenshots of Minecraft, © Mojang AB, shared under Mojang's usage guidelines.
123
- This dataset is not an official Minecraft product and is not approved by or associated with
124
- Mojang or Microsoft.
125
- - Everything derived by the authors (metadata, block maps, shadow traces, manifest) is released
126
- under CC BY 4.0.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ license_name: mixed
4
+ license_details: Annotations CC BY 4.0; images are Minecraft screenshots under Mojang's usage guidelines; Sundial Lite renders made with a GPL-3.0 shader
5
+ pretty_name: Minecraft vanilla frames with depth, camera, traced sun shadows and Sundial Lite renders
6
+ task_categories:
7
+ - image-to-image
8
+ - depth-estimation
9
+ tags:
10
+ - minecraft
11
+ - webdataset
12
+ - depth
13
+ - shadows
14
+ - shaders
15
+ size_categories:
16
+ - 10K<n<100K
17
+ configs:
18
+ - config_name: preview
19
+ default: true
20
+ data_files:
21
+ - split: train
22
+ path: preview/train-*.tar
23
+ - split: test
24
+ path: preview/test-*.tar
25
+ ---
26
+
27
+ # Minecraft vanilla frames with depth, camera, traced sun shadows and Sundial Lite renders
28
+
29
+ 20 787 frames of Minecraft Java Edition 1.21.1 rendered **without shaders** at 1280×720, each with
30
+ its depth buffer, its camera, the block under every pixel and the sun shadow traced in the saved
31
+ world, and for 16 363 of them the **same frame rendered with the Sundial Lite shader pack**. They
32
+ were captured with a Fabric mod that teleports the player through a plan of poses with the world
33
+ frozen (no weather, fixed time of day).
34
+
35
+ Capture tool (the mod, its configuration, the shader options and the contribution validator):
36
+ **<https://github.com/stevenijm777/nrmc-capture>**
37
+
38
+ *Resumen en español: 20 787 fotogramas de Minecraft 1.21.1 sin shaders, con profundidad, cámara,
39
+ bloque por píxel y la sombra del sol trazada en el mundo guardado; 16 363 de ellos también con el
40
+ shader Sundial Lite, para aprender el paso de vanilla a shader.*
41
+
42
+ ## Paired renders: Sundial Lite
43
+
44
+ `sundial0/` holds the target half for the first capture round: the same pose, time and camera
45
+ rendered with **Sundial Lite v1.2.0**
46
+ ([Modrinth](https://modrinth.com/shader/sundial-lite), by geforcelegend, GPL-3.0) through Iris
47
+ 1.8.14. Settings that change from the pack defaults, so that
48
+ the render is a function of the pose alone and the sun follows the same path as in vanilla:
49
+
50
+ ```
51
+ sunPathRotation=0.0 (the pack tilts the sun 40° by default)
52
+ DEPTH_OF_FIELD=false MOTION_BLUR=false AVERAGE_EXPOSURE=false
53
+ VOLUMETRIC_FOG=false RAIN_RIPPLES=false PLANT_WAVE=false CAULDRON_WAVE=false
54
+ CLOUD_SPEED=0.00 PLANE_CLOUD=false PLANE_CLOUD_SPEED=0.00 WATER_WAVE_SPEED=0.0 CLOUD_TYPE=0
55
+ ```
56
+
57
+ Two captures of the same pose differ by about 43 dB PSNR (temporal accumulation inside the pack):
58
+ that is the ceiling for any model of this target. The Sundial renders of the second capture round
59
+ (`-r2` shards) are not captured yet. One scene of the first round has no Sundial render.
60
+
61
+ The renders of Complementary Reimagined used in the accompanying work are not included: its
62
+ licence does not clearly allow redistributing them.
63
+
64
+ ## Splits
65
+
66
+ | Split | Scenes | World |
67
+ |---|---|---|
68
+ | train | 16 706 | `minesr_atlas` |
69
+ | test | 3 381 | `minesr_atlas`, in cells of the map not used for training |
70
+ | test | 700 | `minesr_test2` (swamp), a different world |
71
+
72
+ | Biome | Scenes | of them, second round |
73
+ |---|---|---|
74
+ | forest | 3 990 | |
75
+ | plains | 3 990 | 1 190 |
76
+ | taiga | 3 990 | 644 |
77
+ | desert | 3 989 | 2 589 |
78
+ | swamp | 3 430 | |
79
+ | snowy_plains | 1 398 | |
80
+
81
+ World seeds: `minesr_atlas` 5277617577473417202, `minesr_test2` 7308878374753708357.
82
+
83
+ Every position is captured at seven times of day (`time_of_day`, in game ticks: 0, 1000, 3000,
84
+ 6000, 9000, 11000, 12000). The split is by area: the map is cut into cells of 512×512 blocks and
85
+ each plan of poses belongs to one split.
86
+
87
+ **Second round (October 2026).** Shards named `-r2` hold three new areas: `taiga_2`, `plains_2`
88
+ and `desert_2`. The poses of `taiga_2_test` turned out to lie in the area of the first-round taiga
89
+ training set (4 blocks from a training pose), so they were moved to **train**; the second-round
90
+ test sets are `plains_2_test` and `desert_2_test`. Nothing of the first round changed: same scenes,
91
+ same shards, same split.
92
+
93
+ **Known overlap.** Three cells of the first round hold poses of both splits, near the edges of
94
+ their areas: forest test and swamp train (closest poses 35 blocks apart), plains test and swamp
95
+ train (32 blocks), and snowy plains train and taiga test (182 blocks). They are kept so that
96
+ results on this test set stay comparable with published ones.
97
+
98
+ **Animals.** Removing entities did not work reliably before mod 0.6.1. In the first round 224
99
+ scenes (32 positions) were removed by hand because an animal or a dropped item was in view, and a
100
+ detector based on depth and the saved world still finds an entity in 3 584 of the 16 364 frames
101
+ (0.05 % of the pixels). In the second round, `taiga_2` and `plains_2` were captured with animals
102
+ (mod 0.6.0); `desert_2` with mod 0.6.1, without them.
103
+
104
+ ## Files
105
+
106
+ WebDataset shards, up to 500 scenes and about 1 GB each:
107
+
108
+ - `data/<split>/<world>-<biome>[-r2]-<nnn>.tar`: the vanilla frame and its annotations;
109
+ - `sundial0/<split>/<same name>.tar`: the Sundial Lite render of the scenes of the data shard of
110
+ the same name.
111
+
112
+ `manifest.jsonl` has one line per scene (pose, biome, time, split, `shard` and `sundial0_shard`,
113
+ null when there is no render).
114
+
115
+ The dataset viewer shows `preview/`: a 640×360 JPEG thumbnail and the manifest line of every
116
+ first-round scene. It is only for browsing; the data is in `data/` and `sundial0/`.
117
+
118
+ Inside a data shard, per scene:
119
+
120
+ | File | Content |
121
+ |---|---|
122
+ | `<scene_id>.vanilla.png` | the frame, RGB, 1280×720 |
123
+ | `<scene_id>.depth.npz` | `depth` float32 (720, 1280): OpenGL window-space depth in [0, 1], 1 = far plane or sky, origin top left |
124
+ | `<scene_id>.json` | pose, biome, weather, `time_of_day`, and `camera`: `position`, `view_matrix` and `projection_matrix` (4×4, column major) |
125
+ | `<scene_id>.blocks.npz` | `ids` uint16 (720, 1280): index into `names`, the block under each pixel (`nrmc:sky` for the sky) |
126
+ | `<scene_id>.shadow.npz` | sun shadow of each pixel, traced through the blocks of the saved world (see below) |
127
+
128
+ Inside a `sundial0` shard: `<scene_id>.sundial0.png` (RGB, 1280×720) and `<scene_id>.sundial0.json`
129
+ (the capture metadata of that pass, with the shader options).
130
+
131
+ Linearise the depth with `camera.projection_matrix`; with the view matrix this gives the world
132
+ position of every pixel.
133
+
134
+ ### `shadow.npz`
135
+
136
+ One ray per pixel from its surface towards the sun, through the block grid of the saved world.
137
+ This is information the renderer has and the image alone does not: the occluder can be outside
138
+ the frame.
139
+
140
+ | Array | Content |
141
+ |---|---|
142
+ | `bits` uint8 | flags of what the ray crossed: 1 solid block, 2 leaves, 4 plant; 8 the first occluder projects inside the frame; 16 it is also visible in the frame |
143
+ | `first` float16 | distance in blocks to the first occluder, `inf` if none |
144
+ | `leaves` uint8 | number of leaf cells crossed (capped at 255) |
145
+ | `screen_test` uint8 | internal flag of the tracer |
146
+
147
+ ## Loading
148
+
149
+ Download the shards (`hf download stemjara/minecraft-vanilla-frames --repo-type dataset
150
+ --include "data/*" --include "sundial0/*"`) and read them with `webdataset` or `tarfile`:
151
+
152
+ ```python
153
+ import io, json, numpy as np, webdataset as wds
154
+
155
+ name = "train/minesr_atlas-forest-000.tar"
156
+ renders = {s["__key__"]: s["sundial0.png"] for s in wds.WebDataset(f"sundial0/{name}")}
157
+ for s in wds.WebDataset(f"data/{name}"):
158
+ meta = json.loads(s["json"])
159
+ depth = np.load(io.BytesIO(s["depth.npz"]))["depth"]
160
+ shadow = np.load(io.BytesIO(s["shadow.npz"]))
161
+ vanilla_png, sundial_png = s["vanilla.png"], renders.get(s["__key__"])
162
+ break
163
+ ```
164
+
165
+ ## Licence
166
+
167
+ - The images are screenshots of Minecraft, © Mojang AB, shared under Mojang's usage guidelines.
168
+ This dataset is not an official Minecraft product and is not approved by or associated with
169
+ Mojang or Microsoft.
170
+ - The `sundial0` renders were made with Sundial Lite v1.2.0, a shader pack under GPL-3.0-or-later.
171
+ The licence covers the pack's code, not the images it renders; the pack is credited here, and
172
+ the renders are shared under the same terms as the rest of the images.
173
+ - Everything derived by the authors (metadata, block maps, shadow traces, manifest) is released
174
+ under CC BY 4.0.