HNIVision / examples /145541.json
Mingze's picture
Phase 3.5: method-grid UI + overlays for each method
d41838a verified
Raw
History Blame Contribute Delete
63 kB
{
"merged": {
"human": {
"present": true,
"count": "6-20",
"tags": [
"crowd",
"people",
"person",
"groups",
"families",
"visitors",
"group",
"individual"
],
"evidence": "[detection] YOLO detected 10 'person' object(s) with confidence \u2265 0.5 | [labeling] Google Vision tagged: people, crowd | [segmentation] Detected person in SAM2 masks (8.3% pixel share) | [classification] ResNet classified scene as 'Social and Informal Recreation' (confidence 0.67) | [vlm] Multiple people are visible walking along the path, some in pairs or small groups, and others sitting on benches."
},
"nature": {
"tags": [
"tree",
"park",
"lawn",
"park path",
"woodland"
],
"dominant": "park path",
"evidence": "[labeling] Google Vision tagged: tree | [segmentation] Top natural classes: tree 55.1% | [classification] ResNet classified scene as 'Social and Informal Recreation' (confidence 0.67) | [vlm] The scene is dominated by a wide paved path flanked by large trees forming a canopy overhead, typical of a park path environment.",
"pixel_shares": {
"tree": 0.5508
}
},
"activity": {
"tags": [
"walking",
"picnicking",
"socializing",
"relaxing",
"resting"
],
"dominant": "walking",
"evidence": "[labeling] Google Vision tagged: walking | [classification] ResNet classified scene as 'Social and Informal Recreation' (confidence 0.67) | [vlm] Most people are walking along the path, while a few are sitting on benches along the side."
},
"meaning": {
"tags": [
"recreation",
"leisure",
"social bonding",
"relaxation",
"social interaction",
"aesthetic appreciation"
],
"dominant": "recreation",
"evidence": "[labeling] Google Vision tagged: recreation, leisure | [classification] ResNet classified scene as 'Social and Informal Recreation' (confidence 0.67) | [vlm] The presence of visitors walking, sitting, and engaging in casual social interaction in a green, shaded park setting suggests recreational use."
},
"summary": "The image shows a park path lined with large trees and several people walking or sitting on benches, indicating a natural urban setting with human activity focused on walking and resting.",
"hni_present": "yes",
"hni_strength": "strong",
"confidence": "high",
"source_method": "merged(classification,detection,labeling,segmentation,vlm)",
"raw_output": null
},
"per_method": {
"detection": {
"human": {
"present": true,
"count": "6-10",
"tags": [
"crowd"
],
"evidence": "YOLO detected 10 'person' object(s) with confidence \u2265 0.5"
},
"nature": {
"tags": [],
"dominant": null,
"evidence": null,
"pixel_shares": {}
},
"activity": {
"tags": [],
"dominant": null,
"evidence": null
},
"meaning": {
"tags": [],
"dominant": null,
"evidence": null
},
"summary": null,
"hni_present": null,
"hni_strength": null,
"confidence": null,
"source_method": "detection",
"raw_output": {
"boxes": [
{
"class_name": "person",
"class_id": 0,
"confidence": 0.9291789531707764,
"x1": 208.80508422851562,
"y1": 165.87232971191406,
"x2": 266.8431091308594,
"y2": 340.3775634765625
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.9248374104499817,
"x1": 118.89724731445312,
"y1": 191.29507446289062,
"x2": 178.7018585205078,
"y2": 343.5417175292969
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.8768019676208496,
"x1": 71.72493743896484,
"y1": 208.26809692382812,
"x2": 91.42684173583984,
"y2": 261.8520812988281
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.8498347401618958,
"x1": 94.4805908203125,
"y1": 206.00889587402344,
"x2": 114.3902816772461,
"y2": 260.98748779296875
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.757563054561615,
"x1": 276.9510192871094,
"y1": 195.71615600585938,
"x2": 289.17779541015625,
"y2": 234.41026306152344
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.726512610912323,
"x1": 178.4121856689453,
"y1": 200.0631103515625,
"x2": 195.70655822753906,
"y2": 246.67112731933594
},
{
"class_name": "bench",
"class_id": 13,
"confidence": 0.7018218636512756,
"x1": 0.0,
"y1": 228.548095703125,
"x2": 26.70081329345703,
"y2": 265.36944580078125
},
{
"class_name": "bench",
"class_id": 13,
"confidence": 0.6754437685012817,
"x1": 371.5491027832031,
"y1": 218.84384155273438,
"x2": 398.7369079589844,
"y2": 246.16905212402344
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.638318657875061,
"x1": 330.981689453125,
"y1": 195.75277709960938,
"x2": 347.6373596191406,
"y2": 240.62625122070312
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.6181930899620056,
"x1": 355.6367492675781,
"y1": 206.1461944580078,
"x2": 383.0953063964844,
"y2": 240.19398498535156
},
{
"class_name": "handbag",
"class_id": 26,
"confidence": 0.5667862892150879,
"x1": 143.47824096679688,
"y1": 217.70721435546875,
"x2": 177.73927307128906,
"y2": 273.81109619140625
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.5611206889152527,
"x1": 190.29623413085938,
"y1": 197.56863403320312,
"x2": 205.38731384277344,
"y2": 245.17483520507812
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.5530803203582764,
"x1": 287.9104309082031,
"y1": 193.38514709472656,
"x2": 300.4593811035156,
"y2": 232.68606567382812
},
{
"class_name": "person",
"class_id": 0,
"confidence": 0.4372217357158661,
"x1": 297.4359130859375,
"y1": 198.22799682617188,
"x2": 312.5256042480469,
"y2": 240.25662231445312
},
{
"class_name": "bench",
"class_id": 13,
"confidence": 0.4257430136203766,
"x1": 430.6173095703125,
"y1": 237.9173126220703,
"x2": 499.24237060546875,
"y2": 293.8536376953125
},
{
"class_name": "bench",
"class_id": 13,
"confidence": 0.34134015440940857,
"x1": 371.9604187011719,
"y1": 219.18173217773438,
"x2": 398.608154296875,
"y2": 246.4174041748047
},
{
"class_name": "bench",
"class_id": 13,
"confidence": 0.31105783581733704,
"x1": 430.30035400390625,
"y1": 237.27105712890625,
"x2": 499.7286071777344,
"y2": 301.0824890136719
}
],
"image_size": [
500,
350
]
}
},
"labeling": {
"human": {
"present": true,
"count": null,
"tags": [
"people",
"crowd"
],
"evidence": "Google Vision tagged: people, crowd"
},
"nature": {
"tags": [
"tree"
],
"dominant": "tree",
"evidence": "Google Vision tagged: tree",
"pixel_shares": {}
},
"activity": {
"tags": [
"walking"
],
"dominant": "walking",
"evidence": "Google Vision tagged: walking"
},
"meaning": {
"tags": [
"recreation",
"leisure"
],
"dominant": "recreation",
"evidence": "Google Vision tagged: recreation, leisure"
},
"summary": null,
"hni_present": null,
"hni_strength": null,
"confidence": null,
"source_method": "labeling",
"raw_output": {
"labels": [
{
"description": "People",
"confidence": 0.9943,
"rank_original": 1,
"kept_by_filter": true
},
{
"description": "Tree",
"confidence": 0.9808,
"rank_original": 2,
"kept_by_filter": true
},
{
"description": "Public space",
"confidence": 0.9544,
"rank_original": 3,
"kept_by_filter": true
},
{
"description": "Recreation",
"confidence": 0.9459,
"rank_original": 4,
"kept_by_filter": true
},
{
"description": "Pedestrian",
"confidence": 0.933,
"rank_original": 5,
"kept_by_filter": true
},
{
"description": "Leisure",
"confidence": 0.9051,
"rank_original": 6,
"kept_by_filter": true
},
{
"description": "Vacation",
"confidence": 0.9042,
"rank_original": 7,
"kept_by_filter": false
},
{
"description": "Spring",
"confidence": 0.8943,
"rank_original": 8,
"kept_by_filter": false
},
{
"description": "Travel",
"confidence": 0.805,
"rank_original": 9,
"kept_by_filter": false
},
{
"description": "Walking",
"confidence": 0.7908,
"rank_original": 10,
"kept_by_filter": true
},
{
"description": "Crowd",
"confidence": 0.7704,
"rank_original": 11,
"kept_by_filter": true
},
{
"description": "Shade",
"confidence": 0.7556,
"rank_original": 12,
"kept_by_filter": true
},
{
"description": "Walkway",
"confidence": 0.7494,
"rank_original": 13,
"kept_by_filter": true
},
{
"description": "Sidewalk",
"confidence": 0.6657,
"rank_original": 14,
"kept_by_filter": true
},
{
"description": "Holiday",
"confidence": 0.6386,
"rank_original": 15,
"kept_by_filter": false
},
{
"description": "Park",
"confidence": 0.6236,
"rank_original": 16,
"kept_by_filter": true
},
{
"description": "Path",
"confidence": 0.5088,
"rank_original": 17,
"kept_by_filter": true
}
]
}
},
"segmentation": {
"human": {
"present": true,
"count": null,
"tags": [
"person"
],
"evidence": "Detected person in SAM2 masks (8.3% pixel share)"
},
"nature": {
"tags": [
"tree"
],
"dominant": "tree",
"evidence": "Top natural classes: tree 55.1%",
"pixel_shares": {
"tree": 0.5508
}
},
"activity": {
"tags": [],
"dominant": null,
"evidence": null
},
"meaning": {
"tags": [],
"dominant": null,
"evidence": null
},
"summary": null,
"hni_present": null,
"hni_strength": null,
"confidence": null,
"source_method": "base",
"raw_output": {
"masks": [
{
"mask_id": 0,
"area": 38943,
"pixel_share": 0.22253142857142857,
"bbox": [
0,
216,
499,
133
],
"sam2_iou": 0.9851356148719788,
"sam2_stability": 0.9272955656051636,
"ade20k_label": "sidewalk",
"ade20k_label_idx": 11,
"ade20k_confidence": 0.9696222684436228
},
{
"mask_id": 1,
"area": 7660,
"pixel_share": 0.04377142857142857,
"bbox": [
415,
0,
81,
215
],
"sam2_iou": 0.9397081136703491,
"sam2_stability": 0.9197383522987366,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 2,
"area": 781,
"pixel_share": 0.0044628571428571425,
"bbox": [
0,
0,
35,
50
],
"sam2_iou": 0.9274634122848511,
"sam2_stability": 0.9586465954780579,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 3,
"area": 6073,
"pixel_share": 0.03470285714285714,
"bbox": [
210,
166,
55,
173
],
"sam2_iou": 0.9119857549667358,
"sam2_stability": 0.9414493441581726,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9392392557220484
},
{
"mask_id": 4,
"area": 674,
"pixel_share": 0.0038514285714285716,
"bbox": [
96,
206,
17,
53
],
"sam2_iou": 0.9076976776123047,
"sam2_stability": 0.9664233326911926,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.8011869436201781
},
{
"mask_id": 5,
"area": 4181,
"pixel_share": 0.02389142857142857,
"bbox": [
119,
193,
56,
150
],
"sam2_iou": 0.9054371118545532,
"sam2_stability": 0.8000860810279846,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9296818942836642
},
{
"mask_id": 6,
"area": 7845,
"pixel_share": 0.044828571428571426,
"bbox": [
381,
0,
116,
215
],
"sam2_iou": 0.8990991711616516,
"sam2_stability": 0.9041656851768494,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 7,
"area": 659,
"pixel_share": 0.003765714285714286,
"bbox": [
73,
208,
17,
52
],
"sam2_iou": 0.8917306065559387,
"sam2_stability": 0.9297218322753906,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.8224582701062215
},
{
"mask_id": 8,
"area": 994,
"pixel_share": 0.00568,
"bbox": [
143,
218,
34,
54
],
"sam2_iou": 0.8890727162361145,
"sam2_stability": 0.9341723322868347,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9949698189134809
},
{
"mask_id": 9,
"area": 585,
"pixel_share": 0.0033428571428571426,
"bbox": [
394,
148,
11,
59
],
"sam2_iou": 0.8886787295341492,
"sam2_stability": 0.9546218514442444,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 10,
"area": 266,
"pixel_share": 0.00152,
"bbox": [
225,
312,
14,
26
],
"sam2_iou": 0.8869212865829468,
"sam2_stability": 0.9340659379959106,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.8045112781954887
},
{
"mask_id": 11,
"area": 5145,
"pixel_share": 0.0294,
"bbox": [
376,
211,
123,
88
],
"sam2_iou": 0.8774938583374023,
"sam2_stability": 0.7297389507293701,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.6884353741496598
},
{
"mask_id": 12,
"area": 1742,
"pixel_share": 0.009954285714285714,
"bbox": [
0,
245,
75,
44
],
"sam2_iou": 0.8769538998603821,
"sam2_stability": 0.90021812915802,
"ade20k_label": "sidewalk",
"ade20k_label_idx": 11,
"ade20k_confidence": 0.9678530424799081
},
{
"mask_id": 13,
"area": 421,
"pixel_share": 0.0024057142857142858,
"bbox": [
333,
196,
13,
44
],
"sam2_iou": 0.8746484518051147,
"sam2_stability": 0.9443155527114868,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.5249406175771971
},
{
"mask_id": 14,
"area": 4028,
"pixel_share": 0.023017142857142858,
"bbox": [
383,
207,
116,
57
],
"sam2_iou": 0.8628748655319214,
"sam2_stability": 0.8614136576652527,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.6077457795431976
},
{
"mask_id": 15,
"area": 346,
"pixel_share": 0.001977142857142857,
"bbox": [
191,
197,
13,
46
],
"sam2_iou": 0.8528357148170471,
"sam2_stability": 0.9136490225791931,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.8208092485549133
},
{
"mask_id": 16,
"area": 121,
"pixel_share": 0.0006914285714285715,
"bbox": [
50,
207,
15,
9
],
"sam2_iou": 0.8474244475364685,
"sam2_stability": 0.8897637724876404,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 17,
"area": 1807,
"pixel_share": 0.010325714285714285,
"bbox": [
481,
34,
18,
175
],
"sam2_iou": 0.8420939445495605,
"sam2_stability": 0.8233470916748047,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 18,
"area": 1061,
"pixel_share": 0.006062857142857143,
"bbox": [
481,
78,
18,
71
],
"sam2_iou": 0.8379675149917603,
"sam2_stability": 0.9148745536804199,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 19,
"area": 364,
"pixel_share": 0.00208,
"bbox": [
299,
200,
12,
39
],
"sam2_iou": 0.8331510424613953,
"sam2_stability": 0.8999999761581421,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.7747252747252747
},
{
"mask_id": 20,
"area": 8604,
"pixel_share": 0.049165714285714285,
"bbox": [
295,
0,
202,
215
],
"sam2_iou": 0.8278874158859253,
"sam2_stability": 0.6303341388702393,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 21,
"area": 4813,
"pixel_share": 0.02750285714285714,
"bbox": [
383,
127,
116,
171
],
"sam2_iou": 0.821421205997467,
"sam2_stability": 0.7045103907585144,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.6382713484313318
},
{
"mask_id": 22,
"area": 109,
"pixel_share": 0.0006228571428571429,
"bbox": [
239,
329,
12,
10
],
"sam2_iou": 0.8163496255874634,
"sam2_stability": 0.9459459185600281,
"ade20k_label": "sidewalk",
"ade20k_label_idx": 11,
"ade20k_confidence": 0.5963302752293578
},
{
"mask_id": 23,
"area": 822,
"pixel_share": 0.004697142857142857,
"bbox": [
432,
111,
15,
110
],
"sam2_iou": 0.8079974055290222,
"sam2_stability": 0.9151162505149841,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 0.9732360097323601
},
{
"mask_id": 24,
"area": 4686,
"pixel_share": 0.026777142857142857,
"bbox": [
44,
0,
43,
217
],
"sam2_iou": 0.8065381646156311,
"sam2_stability": 0.8253560066223145,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 25,
"area": 33881,
"pixel_share": 0.1936057142857143,
"bbox": [
150,
0,
348,
214
],
"sam2_iou": 0.8058135509490967,
"sam2_stability": 0.6482624411582947,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 26,
"area": 526,
"pixel_share": 0.0030057142857142856,
"bbox": [
23,
0,
36,
33
],
"sam2_iou": 0.8054989576339722,
"sam2_stability": 0.8886925578117371,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 27,
"area": 506,
"pixel_share": 0.0028914285714285713,
"bbox": [
24,
0,
22,
33
],
"sam2_iou": 0.8032236695289612,
"sam2_stability": 0.9011194109916687,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 28,
"area": 547,
"pixel_share": 0.003125714285714286,
"bbox": [
158,
133,
9,
74
],
"sam2_iou": 0.7977089881896973,
"sam2_stability": 0.8942807912826538,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 29,
"area": 2792,
"pixel_share": 0.015954285714285716,
"bbox": [
0,
32,
51,
94
],
"sam2_iou": 0.7972413897514343,
"sam2_stability": 0.8417001366615295,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 30,
"area": 357,
"pixel_share": 0.00204,
"bbox": [
130,
137,
7,
75
],
"sam2_iou": 0.7875673174858093,
"sam2_stability": 0.9214091897010803,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 31,
"area": 328,
"pixel_share": 0.0018742857142857142,
"bbox": [
358,
206,
24,
32
],
"sam2_iou": 0.7816399335861206,
"sam2_stability": 0.8446327447891235,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9451219512195121
},
{
"mask_id": 32,
"area": 2153,
"pixel_share": 0.012302857142857143,
"bbox": [
0,
219,
72,
41
],
"sam2_iou": 0.7810620069503784,
"sam2_stability": 0.8011265397071838,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.6437529029261495
},
{
"mask_id": 33,
"area": 2630,
"pixel_share": 0.01502857142857143,
"bbox": [
80,
0,
32,
210
],
"sam2_iou": 0.7780913710594177,
"sam2_stability": 0.8614140152931213,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 34,
"area": 6335,
"pixel_share": 0.0362,
"bbox": [
0,
33,
53,
190
],
"sam2_iou": 0.7706394791603088,
"sam2_stability": 0.7414514422416687,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 35,
"area": 281,
"pixel_share": 0.0016057142857142856,
"bbox": [
374,
153,
9,
57
],
"sam2_iou": 0.7697518467903137,
"sam2_stability": 0.8809523582458496,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 0.9572953736654805
},
{
"mask_id": 36,
"area": 1570,
"pixel_share": 0.008971428571428571,
"bbox": [
32,
99,
32,
121
],
"sam2_iou": 0.7690116167068481,
"sam2_stability": 0.7975113391876221,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 37,
"area": 2108,
"pixel_share": 0.012045714285714286,
"bbox": [
414,
7,
40,
207
],
"sam2_iou": 0.759834885597229,
"sam2_stability": 0.790814220905304,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 38,
"area": 338,
"pixel_share": 0.0019314285714285714,
"bbox": [
288,
194,
12,
36
],
"sam2_iou": 0.7588306069374084,
"sam2_stability": 0.8795518279075623,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9201183431952663
},
{
"mask_id": 39,
"area": 5909,
"pixel_share": 0.03376571428571429,
"bbox": [
44,
0,
66,
218
],
"sam2_iou": 0.7515614032745361,
"sam2_stability": 0.6942500472068787,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 40,
"area": 735,
"pixel_share": 0.0042,
"bbox": [
43,
148,
22,
61
],
"sam2_iou": 0.7489481568336487,
"sam2_stability": 0.8421052694320679,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 41,
"area": 518,
"pixel_share": 0.00296,
"bbox": [
371,
219,
26,
26
],
"sam2_iou": 0.7478890419006348,
"sam2_stability": 0.8972477316856384,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.5868725868725869
},
{
"mask_id": 42,
"area": 4777,
"pixel_share": 0.027297142857142857,
"bbox": [
223,
107,
90,
86
],
"sam2_iou": 0.7478458881378174,
"sam2_stability": 0.8135114908218384,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 43,
"area": 584,
"pixel_share": 0.003337142857142857,
"bbox": [
137,
125,
25,
67
],
"sam2_iou": 0.7448092699050903,
"sam2_stability": 0.8616352081298828,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 44,
"area": 193,
"pixel_share": 0.0011028571428571428,
"bbox": [
172,
185,
12,
19
],
"sam2_iou": 0.739743173122406,
"sam2_stability": 0.9396985173225403,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 45,
"area": 25026,
"pixel_share": 0.1430057142857143,
"bbox": [
165,
225,
334,
124
],
"sam2_iou": 0.738775372505188,
"sam2_stability": 0.640962541103363,
"ade20k_label": "sidewalk",
"ade20k_label_idx": 11,
"ade20k_confidence": 1.0
},
{
"mask_id": 46,
"area": 152,
"pixel_share": 0.0008685714285714286,
"bbox": [
187,
172,
9,
30
],
"sam2_iou": 0.7371828556060791,
"sam2_stability": 0.8993710875511169,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 47,
"area": 5715,
"pixel_share": 0.03265714285714286,
"bbox": [
446,
0,
50,
215
],
"sam2_iou": 0.7315586805343628,
"sam2_stability": 0.6971547603607178,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 48,
"area": 272,
"pixel_share": 0.0015542857142857142,
"bbox": [
364,
208,
18,
29
],
"sam2_iou": 0.7263848185539246,
"sam2_stability": 0.8556700944900513,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9889705882352942
},
{
"mask_id": 49,
"area": 160,
"pixel_share": 0.0009142857142857143,
"bbox": [
156,
200,
10,
19
],
"sam2_iou": 0.7243772745132446,
"sam2_stability": 0.910179615020752,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.7375
},
{
"mask_id": 50,
"area": 45,
"pixel_share": 0.00025714285714285715,
"bbox": [
318,
208,
8,
7
],
"sam2_iou": 0.7222965359687805,
"sam2_stability": 0.9130434989929199,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.7777777777777778
},
{
"mask_id": 51,
"area": 700,
"pixel_share": 0.004,
"bbox": [
391,
79,
14,
128
],
"sam2_iou": 0.721119225025177,
"sam2_stability": 0.7506393790245056,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 52,
"area": 379,
"pixel_share": 0.0021657142857142856,
"bbox": [
349,
206,
21,
32
],
"sam2_iou": 0.7176253795623779,
"sam2_stability": 0.7808857560157776,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.783641160949868
},
{
"mask_id": 53,
"area": 56,
"pixel_share": 0.00032,
"bbox": [
205,
207,
6,
9
],
"sam2_iou": 0.7155794501304626,
"sam2_stability": 0.9016393423080444,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 1.0
},
{
"mask_id": 54,
"area": 138,
"pixel_share": 0.0007885714285714285,
"bbox": [
351,
171,
5,
34
],
"sam2_iou": 0.7136868238449097,
"sam2_stability": 0.9241379499435425,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 0.9927536231884058
},
{
"mask_id": 55,
"area": 6984,
"pixel_share": 0.039908571428571425,
"bbox": [
14,
0,
96,
217
],
"sam2_iou": 0.7110123634338379,
"sam2_stability": 0.7437820434570312,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 56,
"area": 184,
"pixel_share": 0.0010514285714285714,
"bbox": [
188,
165,
9,
35
],
"sam2_iou": 0.7083605527877808,
"sam2_stability": 0.8571428656578064,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 57,
"area": 183,
"pixel_share": 0.0010457142857142857,
"bbox": [
310,
215,
22,
9
],
"sam2_iou": 0.7073663473129272,
"sam2_stability": 0.8186274766921997,
"ade20k_label": "sidewalk",
"ade20k_label_idx": 11,
"ade20k_confidence": 0.9890710382513661
},
{
"mask_id": 58,
"area": 276,
"pixel_share": 0.001577142857142857,
"bbox": [
131,
150,
6,
59
],
"sam2_iou": 0.7072437405586243,
"sam2_stability": 0.8316831588745117,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 59,
"area": 656,
"pixel_share": 0.0037485714285714284,
"bbox": [
155,
124,
12,
83
],
"sam2_iou": 0.7064328789710999,
"sam2_stability": 0.7848443984985352,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 60,
"area": 140,
"pixel_share": 0.0008,
"bbox": [
80,
173,
5,
35
],
"sam2_iou": 0.7043730616569519,
"sam2_stability": 0.8571428656578064,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 61,
"area": 3990,
"pixel_share": 0.0228,
"bbox": [
384,
149,
115,
114
],
"sam2_iou": 0.7041664719581604,
"sam2_stability": 0.7635641694068909,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.6140350877192983
},
{
"mask_id": 62,
"area": 192,
"pixel_share": 0.001097142857142857,
"bbox": [
112,
172,
6,
42
],
"sam2_iou": 0.7032967805862427,
"sam2_stability": 0.9187816977500916,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 63,
"area": 135,
"pixel_share": 0.0007714285714285715,
"bbox": [
80,
175,
7,
37
],
"sam2_iou": 0.6978949308395386,
"sam2_stability": 0.8561643958091736,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 64,
"area": 92,
"pixel_share": 0.0005257142857142857,
"bbox": [
0,
182,
9,
25
],
"sam2_iou": 0.6874281764030457,
"sam2_stability": 0.8969072103500366,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 65,
"area": 1814,
"pixel_share": 0.010365714285714285,
"bbox": [
416,
14,
26,
197
],
"sam2_iou": 0.6859089732170105,
"sam2_stability": 0.7718780636787415,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 66,
"area": 124,
"pixel_share": 0.0007085714285714285,
"bbox": [
113,
174,
4,
39
],
"sam2_iou": 0.6776684522628784,
"sam2_stability": 0.8636363744735718,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 67,
"area": 2195,
"pixel_share": 0.012542857142857142,
"bbox": [
223,
133,
63,
69
],
"sam2_iou": 0.6774100661277771,
"sam2_stability": 0.6056337952613831,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 0.9995444191343964
},
{
"mask_id": 68,
"area": 1981,
"pixel_share": 0.01132,
"bbox": [
0,
218,
121,
43
],
"sam2_iou": 0.6727327108383179,
"sam2_stability": 0.7302977442741394,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.6032306915699142
},
{
"mask_id": 69,
"area": 927,
"pixel_share": 0.005297142857142857,
"bbox": [
432,
110,
22,
111
],
"sam2_iou": 0.672470211982727,
"sam2_stability": 0.8637274503707886,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 0.9784250269687162
},
{
"mask_id": 70,
"area": 159,
"pixel_share": 0.0009085714285714286,
"bbox": [
114,
215,
12,
19
],
"sam2_iou": 0.672063946723938,
"sam2_stability": 0.7865168452262878,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.89937106918239
},
{
"mask_id": 71,
"area": 580,
"pixel_share": 0.0033142857142857145,
"bbox": [
0,
229,
25,
34
],
"sam2_iou": 0.6716245412826538,
"sam2_stability": 0.9067103266716003,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.6862068965517242
},
{
"mask_id": 72,
"area": 1297,
"pixel_share": 0.007411428571428571,
"bbox": [
219,
252,
37,
41
],
"sam2_iou": 0.6711581945419312,
"sam2_stability": 0.9546485543251038,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9976869699306091
},
{
"mask_id": 73,
"area": 167,
"pixel_share": 0.0009542857142857143,
"bbox": [
191,
201,
12,
18
],
"sam2_iou": 0.666502058506012,
"sam2_stability": 0.8833333253860474,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9401197604790419
},
{
"mask_id": 74,
"area": 429,
"pixel_share": 0.0024514285714285714,
"bbox": [
138,
133,
20,
60
],
"sam2_iou": 0.6633657217025757,
"sam2_stability": 0.7685950398445129,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 75,
"area": 336,
"pixel_share": 0.00192,
"bbox": [
210,
214,
10,
48
],
"sam2_iou": 0.6597411036491394,
"sam2_stability": 0.8971428275108337,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9285714285714286
},
{
"mask_id": 76,
"area": 637,
"pixel_share": 0.00364,
"bbox": [
0,
265,
70,
30
],
"sam2_iou": 0.6574116349220276,
"sam2_stability": 0.8895522356033325,
"ade20k_label": "sidewalk",
"ade20k_label_idx": 11,
"ade20k_confidence": 0.7676609105180534
},
{
"mask_id": 77,
"area": 200,
"pixel_share": 0.001142857142857143,
"bbox": [
134,
319,
11,
21
],
"sam2_iou": 0.6560770869255066,
"sam2_stability": 0.9061033129692078,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.935
},
{
"mask_id": 78,
"area": 2184,
"pixel_share": 0.01248,
"bbox": [
0,
33,
51,
59
],
"sam2_iou": 0.6465731859207153,
"sam2_stability": 0.7190569639205933,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 79,
"area": 1654,
"pixel_share": 0.00945142857142857,
"bbox": [
226,
138,
53,
54
],
"sam2_iou": 0.6443814635276794,
"sam2_stability": 0.701265811920166,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 80,
"area": 1170,
"pixel_share": 0.006685714285714285,
"bbox": [
428,
108,
25,
142
],
"sam2_iou": 0.642580509185791,
"sam2_stability": 0.6443708539009094,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 0.9213675213675213
},
{
"mask_id": 81,
"area": 3023,
"pixel_share": 0.017274285714285714,
"bbox": [
223,
116,
63,
77
],
"sam2_iou": 0.6415320634841919,
"sam2_stability": 0.7515510320663452,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 82,
"area": 368,
"pixel_share": 0.002102857142857143,
"bbox": [
129,
135,
18,
76
],
"sam2_iou": 0.6294550895690918,
"sam2_stability": 0.8692307472229004,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 83,
"area": 886,
"pixel_share": 0.005062857142857143,
"bbox": [
65,
152,
20,
66
],
"sam2_iou": 0.628684937953949,
"sam2_stability": 0.7310278415679932,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 84,
"area": 27,
"pixel_share": 0.00015428571428571428,
"bbox": [
319,
209,
4,
6
],
"sam2_iou": 0.6260432600975037,
"sam2_stability": 0.931034505367279,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.7407407407407407
},
{
"mask_id": 85,
"area": 448,
"pixel_share": 0.00256,
"bbox": [
124,
104,
29,
107
],
"sam2_iou": 0.6250526905059814,
"sam2_stability": 0.6527777910232544,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 86,
"area": 2248,
"pixel_share": 0.012845714285714285,
"bbox": [
212,
188,
50,
71
],
"sam2_iou": 0.6234187483787537,
"sam2_stability": 0.7845849990844727,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.9839857651245552
},
{
"mask_id": 87,
"area": 1255,
"pixel_share": 0.007171428571428572,
"bbox": [
229,
153,
52,
40
],
"sam2_iou": 0.6227227449417114,
"sam2_stability": 0.7385976910591125,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 88,
"area": 853,
"pixel_share": 0.004874285714285715,
"bbox": [
432,
53,
24,
73
],
"sam2_iou": 0.6180353164672852,
"sam2_stability": 0.7904858589172363,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 89,
"area": 149,
"pixel_share": 0.0008514285714285714,
"bbox": [
173,
188,
10,
15
],
"sam2_iou": 0.6169610023498535,
"sam2_stability": 0.7654321193695068,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 90,
"area": 283,
"pixel_share": 0.0016171428571428571,
"bbox": [
112,
214,
17,
24
],
"sam2_iou": 0.61509770154953,
"sam2_stability": 0.8866666555404663,
"ade20k_label": "person",
"ade20k_label_idx": 12,
"ade20k_confidence": 0.8374558303886925
},
{
"mask_id": 91,
"area": 270,
"pixel_share": 0.001542857142857143,
"bbox": [
488,
266,
11,
34
],
"sam2_iou": 0.6128129363059998,
"sam2_stability": 0.9049295783042908,
"ade20k_label": "fence",
"ade20k_label_idx": 32,
"ade20k_confidence": 0.6333333333333333
},
{
"mask_id": 92,
"area": 3229,
"pixel_share": 0.01845142857142857,
"bbox": [
4,
83,
53,
140
],
"sam2_iou": 0.6027126908302307,
"sam2_stability": 0.6060037612915039,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 93,
"area": 434,
"pixel_share": 0.00248,
"bbox": [
167,
136,
22,
64
],
"sam2_iou": 0.6025635600090027,
"sam2_stability": 0.7148362398147583,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
},
{
"mask_id": 94,
"area": 1132,
"pixel_share": 0.0064685714285714286,
"bbox": [
290,
147,
39,
47
],
"sam2_iou": 0.6012504696846008,
"sam2_stability": 0.7064748406410217,
"ade20k_label": "tree",
"ade20k_label_idx": 4,
"ade20k_confidence": 1.0
}
],
"n_masks": 95,
"image_size": [
500,
350
],
"pixel_shares": {
"shares": {
"tree": 0.5507666889131249,
"sidewalk": 0.2711897514365243,
"fence": 0.09554311200820405,
"person": 0.0825004476421467
},
"image_size": [
500,
350
]
},
"n_classes_detected": 4,
"coverage_ratio": 1.4041828571428572,
"method_meta": {
"sam2_model": "facebook/sam2.1-hiera-small",
"ade20k_model": "nvidia/segformer-b0-finetuned-ade-512-512"
}
}
},
"classification": {
"human": {
"present": true,
"count": null,
"tags": [
"groups",
"families"
],
"evidence": "ResNet classified scene as 'Social and Informal Recreation' (confidence 0.67)"
},
"nature": {
"tags": [
"park",
"lawn"
],
"dominant": "park",
"evidence": "ResNet classified scene as 'Social and Informal Recreation' (confidence 0.67)",
"pixel_shares": {}
},
"activity": {
"tags": [
"picnicking",
"socializing",
"relaxing"
],
"dominant": "picnicking",
"evidence": "ResNet classified scene as 'Social and Informal Recreation' (confidence 0.67)"
},
"meaning": {
"tags": [
"social bonding",
"leisure"
],
"dominant": "social bonding",
"evidence": "ResNet classified scene as 'Social and Informal Recreation' (confidence 0.67)"
},
"summary": null,
"hni_present": null,
"hni_strength": null,
"confidence": "medium",
"source_method": "classification",
"raw_output": {
"predicted_class": "Social and Informal Recreation",
"predicted_class_idx": 2,
"confidence": 0.6697853207588196,
"all_probs": {
"Cultural Performance": 0.0023363404907286167,
"Garden Vegetation": 0.0008488713065162301,
"Social and Informal Recreation": 0.6697853207588196,
"Sport and Exercise Activities": 0.13031455874443054,
"Waterscape and Built Settings": 0.010937741957604885,
"Wildlife Encounters": 0.0005706356023438275,
"Woodland Settings": 0.18520653247833252
}
}
},
"vlm": {
"human": {
"present": true,
"count": "6-20",
"tags": [
"visitors",
"group",
"individual"
],
"evidence": "Multiple people are visible walking along the path, some in pairs or small groups, and others sitting on benches."
},
"nature": {
"tags": [
"park path",
"woodland"
],
"dominant": "park path",
"evidence": "The scene is dominated by a wide paved path flanked by large trees forming a canopy overhead, typical of a park path environment.",
"pixel_shares": {}
},
"activity": {
"tags": [
"walking",
"resting"
],
"dominant": "walking",
"evidence": "Most people are walking along the path, while a few are sitting on benches along the side."
},
"meaning": {
"tags": [
"recreation",
"relaxation",
"social interaction",
"aesthetic appreciation"
],
"dominant": "recreation",
"evidence": "The presence of visitors walking, sitting, and engaging in casual social interaction in a green, shaded park setting suggests recreational use."
},
"summary": "The image shows a park path lined with large trees and several people walking or sitting on benches, indicating a natural urban setting with human activity focused on walking and resting.",
"hni_present": "yes",
"hni_strength": "strong",
"confidence": "high",
"source_method": "vlm",
"raw_output": {
"raw_response": "{\n \"image_level_summary\": \"The image shows a park path lined with large trees and several people walking or sitting on benches, indicating a natural urban setting with human activity focused on walking and resting.\",\n \"human_presence\": {\n \"visible\": \"yes\",\n \"presence_type\": [\"visitors\", \"group\", \"individual\"],\n \"estimated_people_count\": \"6-20\",\n \"visual_evidence\": \"Multiple people are visible walking along the path, some in pairs or small groups, and others sitting on benches.\"\n },\n \"nature_detection\": {\n \"nature_types\": [\"park path\", \"woodland\"],\n \"dominant_nature_type\": \"park path\",\n \"visual_evidence\": \"The scene is dominated by a wide paved path flanked by large trees forming a canopy overhead, typical of a park path environment.\"\n },\n \"activity_evidence\": {\n \"activity_types\": [\"walking\", \"resting\"],\n \"dominant_activity\": \"walking\",\n \"visual_evidence\": \"Most people are walking along the path, while a few are sitting on benches along the side.\"\n },\n \"cultural_experiential_meaning\": {\n \"meaning_types\": [\"recreation\", \"relaxation\", \"social interaction\", \"aesthetic appreciation\"],\n \"dominant_meaning\": \"recreation\",\n \"visual_evidence\": \"The presence of visitors walking, sitting, and engaging in casual social interaction in a green, shaded park setting suggests recreational use.\"\n },\n \"hni_overall_interpretation\": {\n \"is_hni_present\": \"yes\",\n \"hni_strength\": \"strong\",\n \"reason\": \"Clear human presence interacting with a natural park environment through walking and resting activities, indicating strong human-nature interaction.\"\n },\n \"confidence\": \"high\",\n \"uncertainty_notes\": \"None\"\n}",
"parsed": {
"image_level_summary": "The image shows a park path lined with large trees and several people walking or sitting on benches, indicating a natural urban setting with human activity focused on walking and resting.",
"human_presence": {
"visible": "yes",
"presence_type": [
"visitors",
"group",
"individual"
],
"estimated_people_count": "6-20",
"visual_evidence": "Multiple people are visible walking along the path, some in pairs or small groups, and others sitting on benches."
},
"nature_detection": {
"nature_types": [
"park path",
"woodland"
],
"dominant_nature_type": "park path",
"visual_evidence": "The scene is dominated by a wide paved path flanked by large trees forming a canopy overhead, typical of a park path environment."
},
"activity_evidence": {
"activity_types": [
"walking",
"resting"
],
"dominant_activity": "walking",
"visual_evidence": "Most people are walking along the path, while a few are sitting on benches along the side."
},
"cultural_experiential_meaning": {
"meaning_types": [
"recreation",
"relaxation",
"social interaction",
"aesthetic appreciation"
],
"dominant_meaning": "recreation",
"visual_evidence": "The presence of visitors walking, sitting, and engaging in casual social interaction in a green, shaded park setting suggests recreational use."
},
"hni_overall_interpretation": {
"is_hni_present": "yes",
"hni_strength": "strong",
"reason": "Clear human presence interacting with a natural park environment through walking and resting activities, indicating strong human-nature interaction."
},
"confidence": "high",
"uncertainty_notes": "None"
},
"parse_error": null,
"tokens_used": 1379
}
}
}
}