Datasets:
Add rai:hasSyntheticData and RAI 1.0 conformsTo to Croissant metadata
Browse files- metadata.json +128 -42
metadata.json
CHANGED
|
@@ -49,7 +49,10 @@
|
|
| 49 |
"@type": "sc:Dataset",
|
| 50 |
"name": "Substream Recollection",
|
| 51 |
"description": "Substream Recollection is a controlled benchmark for testing whether long-context VLMs / LLMs can recall whether a short probe substream occurred inside a longer parent stream. It pairs synthetic streams (low / medium / uniform entropy, lengths L=8..4096) with EasyHuman rendered 3-belt videos and a natural-video sub-benchmark drawn from EPIC-Kitchens-100 and SoccerNet. The dataset is organized into four top-level configs: text (synthetic substream text-modality questions), synthetic_video (rendered synthetic substream videos), easyhuman (rendered 3-belt EasyHuman videos and their text-modality counterparts), and natural_video (real-world videos).",
|
| 52 |
-
"conformsTo":
|
|
|
|
|
|
|
|
|
|
| 53 |
"citeAs": "Anonymous, \"Substream Recollection,\" 2026 (anonymized for review).",
|
| 54 |
"keywords": [
|
| 55 |
"VLM",
|
|
@@ -74,12 +77,19 @@
|
|
| 74 |
"name": "Anonymous Authors"
|
| 75 |
},
|
| 76 |
"datePublished": "2026-05-03",
|
|
|
|
| 77 |
"rai:dataCollection": "Substream Recollection has two components. (1) Synthetic streams over a fixed alphabet of 16 symbols, generated by a variable-order n-gram process (rule lengths 1-8) and resampled until the realized stream falls within a target empirical Lempel-Ziv entropy band (low ~0.7 bits/symbol, medium ~1.9 bits/symbol, max-entropy = 4.0 bits/symbol = uniform i.i.d.). The 'EasyHuman' diagnostic split is generated separately by hand-specified pattern templates (repeating motifs and counting sequences) with controlled deviations. (2) A natural-video subset of 1,028 clips drawn from EPIC-Kitchens-100 (egocentric cooking videos) and SoccerNet (sports broadcasts) at L in {8, 16, 64, 128, 512, 1024}; EPIC-Kitchens-100 derived clips ship in this repo and SoccerNet rows ship as provenance metadata only.",
|
| 78 |
"rai:dataCollectionType": "Mixed: programmatically generated synthetic streams (text and rendered video) plus curated derivatives of two existing public video datasets (EPIC-Kitchens-100, SoccerNet). No new human data collection was performed.",
|
| 79 |
"rai:dataCollectionRawData": "Synthetic split: no raw data — streams are sampled from an n-gram generator over a 16-symbol alphabet with the seed-and-resample procedure described in the paper (Appendix 'Benchmark Details'). Natural-video split: raw data are EPIC-Kitchens-100 trimmed action clips (CC BY-NC 4.0, redistributed here) and SoccerNet broadcast events (NDA-gated; only provenance pointers are redistributed, not pixels).",
|
| 80 |
"rai:dataCollectionTimeFrame": [
|
| 81 |
-
{
|
| 82 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 83 |
],
|
| 84 |
"rai:dataCollectionMissingData": "None within the synthetic split (the generator is deterministic given a seed). Natural-video coverage is uneven by design: SoccerNet rows are provided as provenance pointers only (NDA-gated source), so users without SoccerNet access cannot run those rows. The natural-video subset is small (1,028 clips), so per-class statistics may be noisy.",
|
| 85 |
"rai:dataAnnotationProtocol": "No human annotators. Synthetic-stream membership labels (yes/no whether a candidate subsequence appears as a contiguous subsequence of the host stream) are derived programmatically from the known generative process: positives are exact substrings sampled from the host stream and negatives are sampled from the same generator (low/medium bands) or i.i.d. uniform (max-entropy band) and rejection-checked against the host stream. EasyHuman labels are derived from the candidate_tag suffix ('_present' / '_absent'). Natural-video labels are derived from EPIC-Kitchens-100 narration metadata and SoccerNet event annotations as released by the upstream projects; no manual relabeling was performed by this work.",
|
|
@@ -1135,8 +1145,12 @@
|
|
| 1135 |
"description": "Stable per-row unique identifier (e.g. 'video1_q0').",
|
| 1136 |
"dataType": "sc:Text",
|
| 1137 |
"source": {
|
| 1138 |
-
"fileObject": {
|
| 1139 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1140 |
}
|
| 1141 |
},
|
| 1142 |
{
|
|
@@ -1146,8 +1160,12 @@
|
|
| 1146 |
"description": "ID of the parent EasyHuman stream/video (e.g. 'len_256_video_001_v0').",
|
| 1147 |
"dataType": "sc:Text",
|
| 1148 |
"source": {
|
| 1149 |
-
"fileObject": {
|
| 1150 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1151 |
}
|
| 1152 |
},
|
| 1153 |
{
|
|
@@ -1157,8 +1175,12 @@
|
|
| 1157 |
"description": "Always 'easyhuman' on this config.",
|
| 1158 |
"dataType": "sc:Text",
|
| 1159 |
"source": {
|
| 1160 |
-
"fileObject": {
|
| 1161 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1162 |
}
|
| 1163 |
},
|
| 1164 |
{
|
|
@@ -1168,8 +1190,12 @@
|
|
| 1168 |
"description": "'text' or 'video'. Text rows ship at L=256 and L=1024; video rows ship at L=256 only.",
|
| 1169 |
"dataType": "sc:Text",
|
| 1170 |
"source": {
|
| 1171 |
-
"fileObject": {
|
| 1172 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1173 |
}
|
| 1174 |
},
|
| 1175 |
{
|
|
@@ -1179,8 +1205,12 @@
|
|
| 1179 |
"description": "Number of items / frames in the parent stream. EasyHuman ships at L in {256, 1024}.",
|
| 1180 |
"dataType": "sc:Integer",
|
| 1181 |
"source": {
|
| 1182 |
-
"fileObject": {
|
| 1183 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1184 |
}
|
| 1185 |
},
|
| 1186 |
{
|
|
@@ -1190,8 +1220,12 @@
|
|
| 1190 |
"description": "Always 'easyhuman' on this config.",
|
| 1191 |
"dataType": "sc:Text",
|
| 1192 |
"source": {
|
| 1193 |
-
"fileObject": {
|
| 1194 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1195 |
}
|
| 1196 |
},
|
| 1197 |
{
|
|
@@ -1201,8 +1235,12 @@
|
|
| 1201 |
"description": "Always 'easyhuman_binary' on this config.",
|
| 1202 |
"dataType": "sc:Text",
|
| 1203 |
"source": {
|
| 1204 |
-
"fileObject": {
|
| 1205 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1206 |
}
|
| 1207 |
},
|
| 1208 |
{
|
|
@@ -1212,8 +1250,12 @@
|
|
| 1212 |
"description": "Natural-language probe shown to the model.",
|
| 1213 |
"dataType": "sc:Text",
|
| 1214 |
"source": {
|
| 1215 |
-
"fileObject": {
|
| 1216 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1217 |
}
|
| 1218 |
},
|
| 1219 |
{
|
|
@@ -1223,8 +1265,12 @@
|
|
| 1223 |
"description": "Ground-truth answer: 'yes' or 'no'.",
|
| 1224 |
"dataType": "sc:Text",
|
| 1225 |
"source": {
|
| 1226 |
-
"fileObject": {
|
| 1227 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1228 |
}
|
| 1229 |
},
|
| 1230 |
{
|
|
@@ -1235,8 +1281,12 @@
|
|
| 1235 |
"dataType": "sc:Text",
|
| 1236 |
"repeated": true,
|
| 1237 |
"source": {
|
| 1238 |
-
"fileObject": {
|
| 1239 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1240 |
}
|
| 1241 |
},
|
| 1242 |
{
|
|
@@ -1246,8 +1296,12 @@
|
|
| 1246 |
"description": "Start time (seconds) of the probe clip within the parent video.",
|
| 1247 |
"dataType": "sc:Float",
|
| 1248 |
"source": {
|
| 1249 |
-
"fileObject": {
|
| 1250 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1251 |
}
|
| 1252 |
},
|
| 1253 |
{
|
|
@@ -1257,8 +1311,12 @@
|
|
| 1257 |
"description": "End time (seconds) of the probe clip within the parent video.",
|
| 1258 |
"dataType": "sc:Float",
|
| 1259 |
"source": {
|
| 1260 |
-
"fileObject": {
|
| 1261 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1262 |
}
|
| 1263 |
},
|
| 1264 |
{
|
|
@@ -1268,8 +1326,12 @@
|
|
| 1268 |
"description": "EasyHuman category tag (e.g. 'x_present', 'mistake_absent').",
|
| 1269 |
"dataType": "sc:Text",
|
| 1270 |
"source": {
|
| 1271 |
-
"fileObject": {
|
| 1272 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1273 |
}
|
| 1274 |
},
|
| 1275 |
{
|
|
@@ -1279,8 +1341,12 @@
|
|
| 1279 |
"description": "Boolean ground-truth: True iff the candidate appears in the parent stream (derived from candidate_tag suffix).",
|
| 1280 |
"dataType": "sc:Boolean",
|
| 1281 |
"source": {
|
| 1282 |
-
"fileObject": {
|
| 1283 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1284 |
}
|
| 1285 |
},
|
| 1286 |
{
|
|
@@ -1290,8 +1356,12 @@
|
|
| 1290 |
"description": "Full text-modality stream rendered as a comma-joined string. Populated on text rows; null on video rows.",
|
| 1291 |
"dataType": "sc:Text",
|
| 1292 |
"source": {
|
| 1293 |
-
"fileObject": {
|
| 1294 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1295 |
}
|
| 1296 |
},
|
| 1297 |
{
|
|
@@ -1301,8 +1371,12 @@
|
|
| 1301 |
"description": "Parent-stream lane track (S_lanes) as a comma-joined string, aligned 1:1 with input_sequence.",
|
| 1302 |
"dataType": "sc:Text",
|
| 1303 |
"source": {
|
| 1304 |
-
"fileObject": {
|
| 1305 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1306 |
}
|
| 1307 |
},
|
| 1308 |
{
|
|
@@ -1312,8 +1386,12 @@
|
|
| 1312 |
"description": "Repo-relative path to the parent EasyHuman video. Populated on video rows; null on text rows.",
|
| 1313 |
"dataType": "sc:Text",
|
| 1314 |
"source": {
|
| 1315 |
-
"fileObject": {
|
| 1316 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1317 |
}
|
| 1318 |
},
|
| 1319 |
{
|
|
@@ -1323,8 +1401,12 @@
|
|
| 1323 |
"description": "Repo-relative path to the EasyHuman probe clip. Populated on video rows; null on text rows.",
|
| 1324 |
"dataType": "sc:Text",
|
| 1325 |
"source": {
|
| 1326 |
-
"fileObject": {
|
| 1327 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1328 |
}
|
| 1329 |
},
|
| 1330 |
{
|
|
@@ -1334,11 +1416,15 @@
|
|
| 1334 |
"description": "Always 'CC-BY-4.0' on EasyHuman.",
|
| 1335 |
"dataType": "sc:Text",
|
| 1336 |
"source": {
|
| 1337 |
-
"fileObject": {
|
| 1338 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1339 |
}
|
| 1340 |
}
|
| 1341 |
]
|
| 1342 |
}
|
| 1343 |
]
|
| 1344 |
-
}
|
|
|
|
| 49 |
"@type": "sc:Dataset",
|
| 50 |
"name": "Substream Recollection",
|
| 51 |
"description": "Substream Recollection is a controlled benchmark for testing whether long-context VLMs / LLMs can recall whether a short probe substream occurred inside a longer parent stream. It pairs synthetic streams (low / medium / uniform entropy, lengths L=8..4096) with EasyHuman rendered 3-belt videos and a natural-video sub-benchmark drawn from EPIC-Kitchens-100 and SoccerNet. The dataset is organized into four top-level configs: text (synthetic substream text-modality questions), synthetic_video (rendered synthetic substream videos), easyhuman (rendered 3-belt EasyHuman videos and their text-modality counterparts), and natural_video (real-world videos).",
|
| 52 |
+
"conformsTo": [
|
| 53 |
+
"http://mlcommons.org/croissant/1.0",
|
| 54 |
+
"http://mlcommons.org/croissant/RAI/1.0"
|
| 55 |
+
],
|
| 56 |
"citeAs": "Anonymous, \"Substream Recollection,\" 2026 (anonymized for review).",
|
| 57 |
"keywords": [
|
| 58 |
"VLM",
|
|
|
|
| 77 |
"name": "Anonymous Authors"
|
| 78 |
},
|
| 79 |
"datePublished": "2026-05-03",
|
| 80 |
+
"rai:hasSyntheticData": true,
|
| 81 |
"rai:dataCollection": "Substream Recollection has two components. (1) Synthetic streams over a fixed alphabet of 16 symbols, generated by a variable-order n-gram process (rule lengths 1-8) and resampled until the realized stream falls within a target empirical Lempel-Ziv entropy band (low ~0.7 bits/symbol, medium ~1.9 bits/symbol, max-entropy = 4.0 bits/symbol = uniform i.i.d.). The 'EasyHuman' diagnostic split is generated separately by hand-specified pattern templates (repeating motifs and counting sequences) with controlled deviations. (2) A natural-video subset of 1,028 clips drawn from EPIC-Kitchens-100 (egocentric cooking videos) and SoccerNet (sports broadcasts) at L in {8, 16, 64, 128, 512, 1024}; EPIC-Kitchens-100 derived clips ship in this repo and SoccerNet rows ship as provenance metadata only.",
|
| 82 |
"rai:dataCollectionType": "Mixed: programmatically generated synthetic streams (text and rendered video) plus curated derivatives of two existing public video datasets (EPIC-Kitchens-100, SoccerNet). No new human data collection was performed.",
|
| 83 |
"rai:dataCollectionRawData": "Synthetic split: no raw data — streams are sampled from an n-gram generator over a 16-symbol alphabet with the seed-and-resample procedure described in the paper (Appendix 'Benchmark Details'). Natural-video split: raw data are EPIC-Kitchens-100 trimmed action clips (CC BY-NC 4.0, redistributed here) and SoccerNet broadcast events (NDA-gated; only provenance pointers are redistributed, not pixels).",
|
| 84 |
"rai:dataCollectionTimeFrame": [
|
| 85 |
+
{
|
| 86 |
+
"@type": "sc:Date",
|
| 87 |
+
"@value": "2025-01-01"
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"@type": "sc:Date",
|
| 91 |
+
"@value": "2026-05-03"
|
| 92 |
+
}
|
| 93 |
],
|
| 94 |
"rai:dataCollectionMissingData": "None within the synthetic split (the generator is deterministic given a seed). Natural-video coverage is uneven by design: SoccerNet rows are provided as provenance pointers only (NDA-gated source), so users without SoccerNet access cannot run those rows. The natural-video subset is small (1,028 clips), so per-class statistics may be noisy.",
|
| 95 |
"rai:dataAnnotationProtocol": "No human annotators. Synthetic-stream membership labels (yes/no whether a candidate subsequence appears as a contiguous subsequence of the host stream) are derived programmatically from the known generative process: positives are exact substrings sampled from the host stream and negatives are sampled from the same generator (low/medium bands) or i.i.d. uniform (max-entropy band) and rejection-checked against the host stream. EasyHuman labels are derived from the candidate_tag suffix ('_present' / '_absent'). Natural-video labels are derived from EPIC-Kitchens-100 narration metadata and SoccerNet event annotations as released by the upstream projects; no manual relabeling was performed by this work.",
|
|
|
|
| 1145 |
"description": "Stable per-row unique identifier (e.g. 'video1_q0').",
|
| 1146 |
"dataType": "sc:Text",
|
| 1147 |
"source": {
|
| 1148 |
+
"fileObject": {
|
| 1149 |
+
"@id": "easyhuman-parquet"
|
| 1150 |
+
},
|
| 1151 |
+
"extract": {
|
| 1152 |
+
"column": "question_id"
|
| 1153 |
+
}
|
| 1154 |
}
|
| 1155 |
},
|
| 1156 |
{
|
|
|
|
| 1160 |
"description": "ID of the parent EasyHuman stream/video (e.g. 'len_256_video_001_v0').",
|
| 1161 |
"dataType": "sc:Text",
|
| 1162 |
"source": {
|
| 1163 |
+
"fileObject": {
|
| 1164 |
+
"@id": "easyhuman-parquet"
|
| 1165 |
+
},
|
| 1166 |
+
"extract": {
|
| 1167 |
+
"column": "stream_id"
|
| 1168 |
+
}
|
| 1169 |
}
|
| 1170 |
},
|
| 1171 |
{
|
|
|
|
| 1175 |
"description": "Always 'easyhuman' on this config.",
|
| 1176 |
"dataType": "sc:Text",
|
| 1177 |
"source": {
|
| 1178 |
+
"fileObject": {
|
| 1179 |
+
"@id": "easyhuman-parquet"
|
| 1180 |
+
},
|
| 1181 |
+
"extract": {
|
| 1182 |
+
"column": "split"
|
| 1183 |
+
}
|
| 1184 |
}
|
| 1185 |
},
|
| 1186 |
{
|
|
|
|
| 1190 |
"description": "'text' or 'video'. Text rows ship at L=256 and L=1024; video rows ship at L=256 only.",
|
| 1191 |
"dataType": "sc:Text",
|
| 1192 |
"source": {
|
| 1193 |
+
"fileObject": {
|
| 1194 |
+
"@id": "easyhuman-parquet"
|
| 1195 |
+
},
|
| 1196 |
+
"extract": {
|
| 1197 |
+
"column": "modality"
|
| 1198 |
+
}
|
| 1199 |
}
|
| 1200 |
},
|
| 1201 |
{
|
|
|
|
| 1205 |
"description": "Number of items / frames in the parent stream. EasyHuman ships at L in {256, 1024}.",
|
| 1206 |
"dataType": "sc:Integer",
|
| 1207 |
"source": {
|
| 1208 |
+
"fileObject": {
|
| 1209 |
+
"@id": "easyhuman-parquet"
|
| 1210 |
+
},
|
| 1211 |
+
"extract": {
|
| 1212 |
+
"column": "length_L"
|
| 1213 |
+
}
|
| 1214 |
}
|
| 1215 |
},
|
| 1216 |
{
|
|
|
|
| 1220 |
"description": "Always 'easyhuman' on this config.",
|
| 1221 |
"dataType": "sc:Text",
|
| 1222 |
"source": {
|
| 1223 |
+
"fileObject": {
|
| 1224 |
+
"@id": "easyhuman-parquet"
|
| 1225 |
+
},
|
| 1226 |
+
"extract": {
|
| 1227 |
+
"column": "entropy_band"
|
| 1228 |
+
}
|
| 1229 |
}
|
| 1230 |
},
|
| 1231 |
{
|
|
|
|
| 1235 |
"description": "Always 'easyhuman_binary' on this config.",
|
| 1236 |
"dataType": "sc:Text",
|
| 1237 |
"source": {
|
| 1238 |
+
"fileObject": {
|
| 1239 |
+
"@id": "easyhuman-parquet"
|
| 1240 |
+
},
|
| 1241 |
+
"extract": {
|
| 1242 |
+
"column": "question_variant"
|
| 1243 |
+
}
|
| 1244 |
}
|
| 1245 |
},
|
| 1246 |
{
|
|
|
|
| 1250 |
"description": "Natural-language probe shown to the model.",
|
| 1251 |
"dataType": "sc:Text",
|
| 1252 |
"source": {
|
| 1253 |
+
"fileObject": {
|
| 1254 |
+
"@id": "easyhuman-parquet"
|
| 1255 |
+
},
|
| 1256 |
+
"extract": {
|
| 1257 |
+
"column": "question_text"
|
| 1258 |
+
}
|
| 1259 |
}
|
| 1260 |
},
|
| 1261 |
{
|
|
|
|
| 1265 |
"description": "Ground-truth answer: 'yes' or 'no'.",
|
| 1266 |
"dataType": "sc:Text",
|
| 1267 |
"source": {
|
| 1268 |
+
"fileObject": {
|
| 1269 |
+
"@id": "easyhuman-parquet"
|
| 1270 |
+
},
|
| 1271 |
+
"extract": {
|
| 1272 |
+
"column": "answer"
|
| 1273 |
+
}
|
| 1274 |
}
|
| 1275 |
},
|
| 1276 |
{
|
|
|
|
| 1281 |
"dataType": "sc:Text",
|
| 1282 |
"repeated": true,
|
| 1283 |
"source": {
|
| 1284 |
+
"fileObject": {
|
| 1285 |
+
"@id": "easyhuman-parquet"
|
| 1286 |
+
},
|
| 1287 |
+
"extract": {
|
| 1288 |
+
"column": "candidate_sequence"
|
| 1289 |
+
}
|
| 1290 |
}
|
| 1291 |
},
|
| 1292 |
{
|
|
|
|
| 1296 |
"description": "Start time (seconds) of the probe clip within the parent video.",
|
| 1297 |
"dataType": "sc:Float",
|
| 1298 |
"source": {
|
| 1299 |
+
"fileObject": {
|
| 1300 |
+
"@id": "easyhuman-parquet"
|
| 1301 |
+
},
|
| 1302 |
+
"extract": {
|
| 1303 |
+
"column": "candidate_clip_start"
|
| 1304 |
+
}
|
| 1305 |
}
|
| 1306 |
},
|
| 1307 |
{
|
|
|
|
| 1311 |
"description": "End time (seconds) of the probe clip within the parent video.",
|
| 1312 |
"dataType": "sc:Float",
|
| 1313 |
"source": {
|
| 1314 |
+
"fileObject": {
|
| 1315 |
+
"@id": "easyhuman-parquet"
|
| 1316 |
+
},
|
| 1317 |
+
"extract": {
|
| 1318 |
+
"column": "candidate_clip_end"
|
| 1319 |
+
}
|
| 1320 |
}
|
| 1321 |
},
|
| 1322 |
{
|
|
|
|
| 1326 |
"description": "EasyHuman category tag (e.g. 'x_present', 'mistake_absent').",
|
| 1327 |
"dataType": "sc:Text",
|
| 1328 |
"source": {
|
| 1329 |
+
"fileObject": {
|
| 1330 |
+
"@id": "easyhuman-parquet"
|
| 1331 |
+
},
|
| 1332 |
+
"extract": {
|
| 1333 |
+
"column": "candidate_tag"
|
| 1334 |
+
}
|
| 1335 |
}
|
| 1336 |
},
|
| 1337 |
{
|
|
|
|
| 1341 |
"description": "Boolean ground-truth: True iff the candidate appears in the parent stream (derived from candidate_tag suffix).",
|
| 1342 |
"dataType": "sc:Boolean",
|
| 1343 |
"source": {
|
| 1344 |
+
"fileObject": {
|
| 1345 |
+
"@id": "easyhuman-parquet"
|
| 1346 |
+
},
|
| 1347 |
+
"extract": {
|
| 1348 |
+
"column": "candidate_present"
|
| 1349 |
+
}
|
| 1350 |
}
|
| 1351 |
},
|
| 1352 |
{
|
|
|
|
| 1356 |
"description": "Full text-modality stream rendered as a comma-joined string. Populated on text rows; null on video rows.",
|
| 1357 |
"dataType": "sc:Text",
|
| 1358 |
"source": {
|
| 1359 |
+
"fileObject": {
|
| 1360 |
+
"@id": "easyhuman-parquet"
|
| 1361 |
+
},
|
| 1362 |
+
"extract": {
|
| 1363 |
+
"column": "input_sequence"
|
| 1364 |
+
}
|
| 1365 |
}
|
| 1366 |
},
|
| 1367 |
{
|
|
|
|
| 1371 |
"description": "Parent-stream lane track (S_lanes) as a comma-joined string, aligned 1:1 with input_sequence.",
|
| 1372 |
"dataType": "sc:Text",
|
| 1373 |
"source": {
|
| 1374 |
+
"fileObject": {
|
| 1375 |
+
"@id": "easyhuman-parquet"
|
| 1376 |
+
},
|
| 1377 |
+
"extract": {
|
| 1378 |
+
"column": "input_sequence_lanes"
|
| 1379 |
+
}
|
| 1380 |
}
|
| 1381 |
},
|
| 1382 |
{
|
|
|
|
| 1386 |
"description": "Repo-relative path to the parent EasyHuman video. Populated on video rows; null on text rows.",
|
| 1387 |
"dataType": "sc:Text",
|
| 1388 |
"source": {
|
| 1389 |
+
"fileObject": {
|
| 1390 |
+
"@id": "easyhuman-parquet"
|
| 1391 |
+
},
|
| 1392 |
+
"extract": {
|
| 1393 |
+
"column": "video_path"
|
| 1394 |
+
}
|
| 1395 |
}
|
| 1396 |
},
|
| 1397 |
{
|
|
|
|
| 1401 |
"description": "Repo-relative path to the EasyHuman probe clip. Populated on video rows; null on text rows.",
|
| 1402 |
"dataType": "sc:Text",
|
| 1403 |
"source": {
|
| 1404 |
+
"fileObject": {
|
| 1405 |
+
"@id": "easyhuman-parquet"
|
| 1406 |
+
},
|
| 1407 |
+
"extract": {
|
| 1408 |
+
"column": "clip_path"
|
| 1409 |
+
}
|
| 1410 |
}
|
| 1411 |
},
|
| 1412 |
{
|
|
|
|
| 1416 |
"description": "Always 'CC-BY-4.0' on EasyHuman.",
|
| 1417 |
"dataType": "sc:Text",
|
| 1418 |
"source": {
|
| 1419 |
+
"fileObject": {
|
| 1420 |
+
"@id": "easyhuman-parquet"
|
| 1421 |
+
},
|
| 1422 |
+
"extract": {
|
| 1423 |
+
"column": "license"
|
| 1424 |
+
}
|
| 1425 |
}
|
| 1426 |
}
|
| 1427 |
]
|
| 1428 |
}
|
| 1429 |
]
|
| 1430 |
+
}
|