{ "merged": { "human": { "present": true, "count": "2-5", "tags": [ "groups", "people in nature", "person", "children" ], "evidence": "[detection] YOLO detected 2 'person' object(s) with confidence \u2265 0.5 | [labeling] Google Vision tagged: people in nature | [segmentation] Detected person in SAM2 masks (1.4% pixel share) | [vlm] Two children are visible standing side by side facing the waterfront." }, "nature": { "tags": [ "cloud", "cumulus", "meteorological phenomenon", "park", "sky", "tree", "grass", "garden", "flowers", "plants", "manicured vegetation", "waterfront", "lawn", "woodland" ], "dominant": "waterfront", "evidence": "[labeling] Google Vision tagged: cloud, cumulus, meteorological phenomenon, park | [segmentation] Top natural classes: sky 43.2%, tree 18.8%, grass 11.5% | [classification] ResNet classified scene as 'Garden Vegetation' (confidence 0.61) | [vlm] A large body of water is visible beyond the lawn and trees, with a grassy area and scattered trees in the foreground.", "pixel_shares": { "sky": 0.432, "tree": 0.1879, "grass": 0.1147 } }, "activity": { "tags": [ "walking", "watching", "landscape viewing" ], "dominant": "landscape viewing", "evidence": "[labeling] Google Vision tagged: walking | [vlm] The children are standing still and looking out over the water and landscape." }, "meaning": { "tags": [ "leisure", "aesthetic appreciation", "recreation", "relaxation" ], "dominant": "aesthetic appreciation", "evidence": "[labeling] Google Vision tagged: leisure | [classification] ResNet classified scene as 'Garden Vegetation' (confidence 0.61) | [vlm] The scene suggests appreciation of the natural landscape and a recreational outdoor setting." }, "summary": "Two children stand by a stone balustrade overlooking a waterfront with trees and lawn under a partly cloudy sky.", "hni_present": "yes", "hni_strength": "moderate", "confidence": "high", "source_method": "merged(classification,detection,labeling,segmentation,vlm)", "raw_output": null }, "per_method": { "detection": { "human": { "present": true, "count": "2-5", "tags": [ "groups" ], "evidence": "YOLO detected 2 '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": 0.8968237042427063, "x1": 264.1493225097656, "y1": 230.96568298339844, "x2": 300.9723205566406, "y2": 342.3010559082031 }, { "class_name": "person", "class_id": 0, "confidence": 0.7865123748779297, "x1": 236.20135498046875, "y1": 237.9204864501953, "x2": 265.6833801269531, "y2": 339.0294189453125 } ], "image_size": [ 524, 349 ] } }, "labeling": { "human": { "present": true, "count": null, "tags": [ "people in nature" ], "evidence": "Google Vision tagged: people in nature" }, "nature": { "tags": [ "cloud", "cumulus", "meteorological phenomenon", "park" ], "dominant": "cloud", "evidence": "Google Vision tagged: cloud, cumulus, meteorological phenomenon, park", "pixel_shares": {} }, "activity": { "tags": [ "walking" ], "dominant": "walking", "evidence": "Google Vision tagged: walking" }, "meaning": { "tags": [ "leisure" ], "dominant": "leisure", "evidence": "Google Vision tagged: leisure" }, "summary": null, "hni_present": null, "hni_strength": null, "confidence": null, "source_method": "labeling", "raw_output": { "labels": [ { "description": "Cloud", "confidence": 0.9201, "rank_original": 1, "kept_by_filter": true }, { "description": "Leisure", "confidence": 0.8528, "rank_original": 2, "kept_by_filter": true }, { "description": "People in nature", "confidence": 0.8367, "rank_original": 3, "kept_by_filter": true }, { "description": "Travel", "confidence": 0.7441, "rank_original": 4, "kept_by_filter": false }, { "description": "Vacation", "confidence": 0.7152, "rank_original": 5, "kept_by_filter": false }, { "description": "Cumulus", "confidence": 0.6656, "rank_original": 6, "kept_by_filter": true }, { "description": "Walkway", "confidence": 0.6043, "rank_original": 7, "kept_by_filter": true }, { "description": "Meteorological phenomenon", "confidence": 0.5979, "rank_original": 8, "kept_by_filter": true }, { "description": "Fence", "confidence": 0.5839, "rank_original": 9, "kept_by_filter": true }, { "description": "Park", "confidence": 0.5821, "rank_original": 10, "kept_by_filter": true }, { "description": "Lake District", "confidence": 0.5689, "rank_original": 11, "kept_by_filter": true }, { "description": "Walking", "confidence": 0.5398, "rank_original": 12, "kept_by_filter": true }, { "description": "Holiday", "confidence": 0.5239, "rank_original": 13, "kept_by_filter": false } ] } }, "segmentation": { "human": { "present": true, "count": null, "tags": [ "person" ], "evidence": "Detected person in SAM2 masks (1.4% pixel share)" }, "nature": { "tags": [ "sky", "tree", "grass" ], "dominant": "sky", "evidence": "Top natural classes: sky 43.2%, tree 18.8%, grass 11.5%", "pixel_shares": { "sky": 0.432, "tree": 0.1879, "grass": 0.1147 } }, "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": 90695, "pixel_share": 0.49593713773267134, "bbox": [ 0, 0, 523, 190 ], "sam2_iou": 0.9943562150001526, "sam2_stability": 0.9858161211013794, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 1, "area": 3206, "pixel_share": 0.017531004615149063, "bbox": [ 32, 92, 99, 44 ], "sam2_iou": 0.9559070467948914, "sam2_stability": 0.9476248621940613, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 2, "area": 1197, "pixel_share": 0.006545418753691025, "bbox": [ 135, 98, 58, 35 ], "sam2_iou": 0.929534375667572, "sam2_stability": 0.9502447247505188, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 3, "area": 14791, "pixel_share": 0.08087994050613531, "bbox": [ 0, 0, 361, 76 ], "sam2_iou": 0.9295023083686829, "sam2_stability": 0.9225113987922668, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 4, "area": 4851, "pixel_share": 0.026526170738642576, "bbox": [ 0, 186, 306, 22 ], "sam2_iou": 0.9225348830223083, "sam2_stability": 0.9319579601287842, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.9029066171923315 }, { "mask_id": 5, "area": 4896, "pixel_share": 0.02677223911284149, "bbox": [ 0, 187, 306, 38 ], "sam2_iou": 0.912516176700592, "sam2_stability": 0.9351979494094849, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.9058415032679739 }, { "mask_id": 6, "area": 1837, "pixel_share": 0.010045057853408867, "bbox": [ 236, 243, 26, 95 ], "sam2_iou": 0.9122116565704346, "sam2_stability": 0.9283456206321716, "ade20k_label": "person", "ade20k_label_idx": 12, "ade20k_confidence": 0.9972781709308656 }, { "mask_id": 7, "area": 7503, "pixel_share": 0.041027800258098386, "bbox": [ 0, 229, 523, 34 ], "sam2_iou": 0.9087814688682556, "sam2_stability": 0.852580189704895, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.9155004664800747 }, { "mask_id": 8, "area": 2312, "pixel_share": 0.012642446247730702, "bbox": [ 263, 232, 32, 108 ], "sam2_iou": 0.9025468230247498, "sam2_stability": 0.9256644248962402, "ade20k_label": "person", "ade20k_label_idx": 12, "ade20k_confidence": 0.9788062283737025 }, { "mask_id": 9, "area": 4958, "pixel_share": 0.027111266650626655, "bbox": [ 0, 186, 523, 45 ], "sam2_iou": 0.8917723298072815, "sam2_stability": 0.6477987170219421, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.8979427188382412 }, { "mask_id": 10, "area": 7964, "pixel_share": 0.04354863404711389, "bbox": [ 309, 127, 130, 107 ], "sam2_iou": 0.8834188580513, "sam2_stability": 0.7927767038345337, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.9702410848819688 }, { "mask_id": 11, "area": 1107, "pixel_share": 0.006053282005293204, "bbox": [ 443, 227, 53, 34 ], "sam2_iou": 0.8820576071739197, "sam2_stability": 0.9170984625816345, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.982836495031617 }, { "mask_id": 12, "area": 7378, "pixel_share": 0.040344276996434746, "bbox": [ 0, 230, 523, 66 ], "sam2_iou": 0.8817605972290039, "sam2_stability": 0.7068063020706177, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.8934670642450528 }, { "mask_id": 13, "area": 1807, "pixel_share": 0.009881012270609594, "bbox": [ 234, 192, 60, 41 ], "sam2_iou": 0.8795121312141418, "sam2_stability": 0.9263771176338196, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 14, "area": 284, "pixel_share": 0.0015529648504997921, "bbox": [ 15, 219, 21, 16 ], "sam2_iou": 0.8744196891784668, "sam2_stability": 0.9584774971008301, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 15, "area": 7259, "pixel_share": 0.03969356285133096, "bbox": [ 0, 239, 521, 27 ], "sam2_iou": 0.8738710284233093, "sam2_stability": 0.84097820520401, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.9370436699269872 }, { "mask_id": 16, "area": 7140, "pixel_share": 0.03904284870622717, "bbox": [ 0, 239, 521, 46 ], "sam2_iou": 0.8721606135368347, "sam2_stability": 0.8608260750770569, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.9495798319327731 }, { "mask_id": 17, "area": 22034, "pixel_share": 0.12048601237997332, "bbox": [ 0, 262, 523, 66 ], "sam2_iou": 0.8698945641517639, "sam2_stability": 0.6539778113365173, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 0.5420713442861033 }, { "mask_id": 18, "area": 227, "pixel_share": 0.001241278243181172, "bbox": [ 514, 274, 9, 27 ], "sam2_iou": 0.869159460067749, "sam2_stability": 0.9563318490982056, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 0.947136563876652 }, { "mask_id": 19, "area": 718, "pixel_share": 0.003926157614995954, "bbox": [ 199, 215, 51, 18 ], "sam2_iou": 0.8673715591430664, "sam2_stability": 0.9176626801490784, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.9832869080779945 }, { "mask_id": 20, "area": 3330, "pixel_share": 0.018209059690719395, "bbox": [ 32, 91, 157, 45 ], "sam2_iou": 0.8631260395050049, "sam2_stability": 0.7742092609405518, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 21, "area": 8539, "pixel_share": 0.04669284105076664, "bbox": [ 0, 330, 523, 18 ], "sam2_iou": 0.8556793332099915, "sam2_stability": 0.8579527139663696, "ade20k_label": "sidewalk", "ade20k_label_idx": 11, "ade20k_confidence": 0.33774446656517154 }, { "mask_id": 22, "area": 13091, "pixel_share": 0.07158402414750979, "bbox": [ 0, 318, 523, 30 ], "sam2_iou": 0.8383881449699402, "sam2_stability": 0.7271488904953003, "ade20k_label": "sidewalk", "ade20k_label_idx": 11, "ade20k_confidence": 0.36009472156443356 }, { "mask_id": 23, "area": 317, "pixel_share": 0.0017334149915789934, "bbox": [ 108, 213, 20, 19 ], "sam2_iou": 0.8299145698547363, "sam2_stability": 0.9184290170669556, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 24, "area": 293, "pixel_share": 0.0016021785253395743, "bbox": [ 419, 127, 27, 14 ], "sam2_iou": 0.8238321542739868, "sam2_stability": 0.9177631735801697, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 25, "area": 1642, "pixel_share": 0.008978761565213588, "bbox": [ 487, 171, 36, 60 ], "sam2_iou": 0.821794331073761, "sam2_stability": 0.9051774144172668, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 26, "area": 264, "pixel_share": 0.0014436011286336097, "bbox": [ 70, 273, 10, 29 ], "sam2_iou": 0.8168237805366516, "sam2_stability": 0.8892857432365417, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 1.0 }, { "mask_id": 27, "area": 90, "pixel_share": 0.0004921367483978215, "bbox": [ 52, 239, 9, 10 ], "sam2_iou": 0.8115766644477844, "sam2_stability": 0.945652186870575, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.6888888888888889 }, { "mask_id": 28, "area": 285, "pixel_share": 0.0015584330365931013, "bbox": [ 163, 228, 29, 10 ], "sam2_iou": 0.8107022047042847, "sam2_stability": 0.9484536051750183, "ade20k_label": "fence", "ade20k_label_idx": 32, "ade20k_confidence": 0.3684210526315789 }, { "mask_id": 29, "area": 435, "pixel_share": 0.0023786609505894703, "bbox": [ 250, 0, 35, 16 ], "sam2_iou": 0.8094030618667603, "sam2_stability": 0.9377777576446533, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 30, "area": 2635, "pixel_share": 0.014408670355869551, "bbox": [ 202, 192, 102, 42 ], "sam2_iou": 0.8078343868255615, "sam2_stability": 0.6968516707420349, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 31, "area": 12205, "pixel_share": 0.0667392112688379, "bbox": [ 0, 304, 523, 44 ], "sam2_iou": 0.8076982498168945, "sam2_stability": 0.6280649304389954, "ade20k_label": "sidewalk", "ade20k_label_idx": 11, "ade20k_confidence": 0.33396149119213436 }, { "mask_id": 32, "area": 65, "pixel_share": 0.0003554320960650933, "bbox": [ 250, 328, 7, 9 ], "sam2_iou": 0.803896427154541, "sam2_stability": 0.9264705777168274, "ade20k_label": "person", "ade20k_label_idx": 12, "ade20k_confidence": 1.0 }, { "mask_id": 33, "area": 324, "pixel_share": 0.0017716922942321574, "bbox": [ 497, 274, 15, 27 ], "sam2_iou": 0.7935760021209717, "sam2_stability": 0.8941176533699036, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 1.0 }, { "mask_id": 34, "area": 213, "pixel_share": 0.0011647236378748442, "bbox": [ 337, 307, 18, 12 ], "sam2_iou": 0.7891997694969177, "sam2_stability": 0.9140271544456482, "ade20k_label": "wall", "ade20k_label_idx": 0, "ade20k_confidence": 0.6901408450704225 }, { "mask_id": 35, "area": 186, "pixel_share": 0.0010170826133554978, "bbox": [ 263, 168, 20, 11 ], "sam2_iou": 0.7859196662902832, "sam2_stability": 0.9025641083717346, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 36, "area": 295, "pixel_share": 0.0016131148975261926, "bbox": [ 481, 274, 14, 27 ], "sam2_iou": 0.7783059477806091, "sam2_stability": 0.933993399143219, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 1.0 }, { "mask_id": 37, "area": 4993, "pixel_share": 0.027302653163892472, "bbox": [ 433, 166, 90, 67 ], "sam2_iou": 0.7765355706214905, "sam2_stability": 0.8762788772583008, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 38, "area": 163, "pixel_share": 0.0008913143332093877, "bbox": [ 372, 232, 15, 14 ], "sam2_iou": 0.7700144052505493, "sam2_stability": 0.9464285969734192, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.6134969325153374 }, { "mask_id": 39, "area": 1335, "pixel_share": 0.007300028434567685, "bbox": [ 0, 208, 242, 22 ], "sam2_iou": 0.7696083784103394, "sam2_stability": 0.8282417058944702, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.700374531835206 }, { "mask_id": 40, "area": 204, "pixel_share": 0.001115509963035062, "bbox": [ 43, 219, 16, 16 ], "sam2_iou": 0.7690678834915161, "sam2_stability": 0.8853210806846619, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 41, "area": 172, "pixel_share": 0.0009405280080491699, "bbox": [ 209, 307, 15, 11 ], "sam2_iou": 0.7686994671821594, "sam2_stability": 0.8994413614273071, "ade20k_label": "wall", "ade20k_label_idx": 0, "ade20k_confidence": 0.7965116279069767 }, { "mask_id": 42, "area": 260, "pixel_share": 0.0014217283842603732, "bbox": [ 101, 272, 10, 28 ], "sam2_iou": 0.765641987323761, "sam2_stability": 0.9032257795333862, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 1.0 }, { "mask_id": 43, "area": 447, "pixel_share": 0.00244427918370918, "bbox": [ 250, 0, 64, 17 ], "sam2_iou": 0.7627469897270203, "sam2_stability": 0.9308855533599854, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 44, "area": 156, "pixel_share": 0.0008530370305562238, "bbox": [ 120, 238, 16, 10 ], "sam2_iou": 0.7482675313949585, "sam2_stability": 0.9329268336296082, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.967948717948718 }, { "mask_id": 45, "area": 83, "pixel_share": 0.0004538594457446576, "bbox": [ 515, 139, 8, 12 ], "sam2_iou": 0.7402138710021973, "sam2_stability": 0.8764045238494873, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 46, "area": 237, "pixel_share": 0.0012959601041142633, "bbox": [ 87, 272, 8, 28 ], "sam2_iou": 0.7217749953269958, "sam2_stability": 0.9306122660636902, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 1.0 }, { "mask_id": 47, "area": 11173, "pixel_share": 0.06109604322054288, "bbox": [ 299, 127, 208, 107 ], "sam2_iou": 0.720637857913971, "sam2_stability": 0.6599676609039307, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.9781616396670545 }, { "mask_id": 48, "area": 230, "pixel_share": 0.0012576828014610993, "bbox": [ 117, 271, 10, 28 ], "sam2_iou": 0.7163692712783813, "sam2_stability": 0.902953565120697, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 1.0 }, { "mask_id": 49, "area": 159, "pixel_share": 0.0008694415888361513, "bbox": [ 70, 0, 32, 5 ], "sam2_iou": 0.7159649133682251, "sam2_stability": 0.9506173133850098, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 50, "area": 118, "pixel_share": 0.0006452459590104771, "bbox": [ 0, 228, 14, 7 ], "sam2_iou": 0.7156550288200378, "sam2_stability": 0.9200000166893005, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 51, "area": 30, "pixel_share": 0.00016404558279927383, "bbox": [ 135, 329, 8, 4 ], "sam2_iou": 0.7005093693733215, "sam2_stability": 0.7878788113594055, "ade20k_label": "sidewalk", "ade20k_label_idx": 11, "ade20k_confidence": 1.0 }, { "mask_id": 52, "area": 2678, "pixel_share": 0.014643802357881843, "bbox": [ 0, 300, 199, 17 ], "sam2_iou": 0.6889805793762207, "sam2_stability": 0.8039411902427673, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 0.38722927557879017 }, { "mask_id": 53, "area": 240, "pixel_share": 0.0013123646623941906, "bbox": [ 133, 271, 9, 28 ], "sam2_iou": 0.6791048049926758, "sam2_stability": 0.9039999842643738, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 1.0 }, { "mask_id": 54, "area": 93, "pixel_share": 0.0005085413066777489, "bbox": [ 12, 239, 11, 9 ], "sam2_iou": 0.6755639314651489, "sam2_stability": 0.9052631855010986, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.5698924731182796 }, { "mask_id": 55, "area": 2286, "pixel_share": 0.012500273409304666, "bbox": [ 234, 192, 84, 42 ], "sam2_iou": 0.6730173826217651, "sam2_stability": 0.77579665184021, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 56, "area": 14792, "pixel_share": 0.08088540869222861, "bbox": [ 0, 0, 523, 76 ], "sam2_iou": 0.6628779768943787, "sam2_stability": 0.908751368522644, "ade20k_label": "sky", "ade20k_label_idx": 2, "ade20k_confidence": 1.0 }, { "mask_id": 57, "area": 1098, "pixel_share": 0.006004068330453422, "bbox": [ 143, 208, 57, 21 ], "sam2_iou": 0.6563957929611206, "sam2_stability": 0.8748928904533386, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.9972677595628415 }, { "mask_id": 58, "area": 1853, "pixel_share": 0.010132548830901813, "bbox": [ 0, 229, 367, 16 ], "sam2_iou": 0.652439296245575, "sam2_stability": 0.6698858737945557, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.43335132218024824 }, { "mask_id": 59, "area": 287, "pixel_share": 0.0015693694087797196, "bbox": [ 449, 274, 13, 26 ], "sam2_iou": 0.648118257522583, "sam2_stability": 0.920265793800354, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 1.0 }, { "mask_id": 60, "area": 4640, "pixel_share": 0.02537238347295435, "bbox": [ 0, 239, 387, 23 ], "sam2_iou": 0.6448993682861328, "sam2_stability": 0.6338375806808472, "ade20k_label": "grass", "ade20k_label_idx": 9, "ade20k_confidence": 0.9672413793103448 }, { "mask_id": 61, "area": 11177, "pixel_share": 0.06111791596491612, "bbox": [ 0, 261, 523, 44 ], "sam2_iou": 0.6447466611862183, "sam2_stability": 0.6076890826225281, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 0.5727833944707882 }, { "mask_id": 62, "area": 1069, "pixel_share": 0.005845490933747457, "bbox": [ 0, 207, 236, 16 ], "sam2_iou": 0.6314085721969604, "sam2_stability": 0.6805877685546875, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 0.6314312441534145 }, { "mask_id": 63, "area": 2653, "pixel_share": 0.014507097705549116, "bbox": [ 0, 300, 523, 20 ], "sam2_iou": 0.6287875771522522, "sam2_stability": 0.7006678581237793, "ade20k_label": "bannister", "ade20k_label_idx": 95, "ade20k_confidence": 0.3660007538635507 }, { "mask_id": 64, "area": 828, "pixel_share": 0.0045276580852599575, "bbox": [ 233, 192, 59, 19 ], "sam2_iou": 0.6138895750045776, "sam2_stability": 0.9034482836723328, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 }, { "mask_id": 65, "area": 235, "pixel_share": 0.001285023731927645, "bbox": [ 82, 208, 14, 26 ], "sam2_iou": 0.6103206872940063, "sam2_stability": 0.8893280625343323, "ade20k_label": "tree", "ade20k_label_idx": 4, "ade20k_confidence": 1.0 } ], "n_masks": 66, "image_size": [ 524, 349 ], "pixel_shares": { "shares": { "sky": 0.43195739547160256, "tree": 0.18786450800167961, "bannister": 0.13632515946704343, "grass": 0.11471629196632693, "sidewalk": 0.11286000893148747, "person": 0.01404376428871366, "wall": 0.0012830681659123779, "fence": 0.000949803707233838 }, "image_size": [ 524, 349 ] }, "n_classes_detected": 8, "coverage_ratio": 1.6407948555305234, "method_meta": { "sam2_model": "facebook/sam2.1-hiera-small", "ade20k_model": "nvidia/segformer-b0-finetuned-ade-512-512" } } }, "classification": { "human": { "present": false, "count": null, "tags": [], "evidence": null }, "nature": { "tags": [ "garden", "flowers", "plants", "manicured vegetation" ], "dominant": "garden", "evidence": "ResNet classified scene as 'Garden Vegetation' (confidence 0.61)", "pixel_shares": {} }, "activity": { "tags": [], "dominant": null, "evidence": null }, "meaning": { "tags": [ "aesthetic appreciation" ], "dominant": "aesthetic appreciation", "evidence": "ResNet classified scene as 'Garden Vegetation' (confidence 0.61)" }, "summary": null, "hni_present": null, "hni_strength": null, "confidence": "medium", "source_method": "classification", "raw_output": { "predicted_class": "Garden Vegetation", "predicted_class_idx": 1, "confidence": 0.6078799366950989, "all_probs": { "Cultural Performance": 0.02940448373556137, "Garden Vegetation": 0.6078799366950989, "Social and Informal Recreation": 0.003936419729143381, "Sport and Exercise Activities": 0.0021129236556589603, "Waterscape and Built Settings": 0.22470790147781372, "Wildlife Encounters": 0.0014671219978481531, "Woodland Settings": 0.1304912567138672 } } }, "vlm": { "human": { "present": true, "count": "2-5", "tags": [ "children" ], "evidence": "Two children are visible standing side by side facing the waterfront." }, "nature": { "tags": [ "waterfront", "lawn", "woodland" ], "dominant": "waterfront", "evidence": "A large body of water is visible beyond the lawn and trees, with a grassy area and scattered trees in the foreground.", "pixel_shares": {} }, "activity": { "tags": [ "watching", "landscape viewing" ], "dominant": "landscape viewing", "evidence": "The children are standing still and looking out over the water and landscape." }, "meaning": { "tags": [ "aesthetic appreciation", "recreation", "relaxation" ], "dominant": "aesthetic appreciation", "evidence": "The scene suggests appreciation of the natural landscape and a recreational outdoor setting." }, "summary": "Two children stand by a stone balustrade overlooking a waterfront with trees and lawn under a partly cloudy sky.", "hni_present": "yes", "hni_strength": "moderate", "confidence": "high", "source_method": "vlm", "raw_output": { "raw_response": "{\n \"image_level_summary\": \"Two children stand by a stone balustrade overlooking a waterfront with trees and lawn under a partly cloudy sky.\",\n \"human_presence\": {\n \"visible\": \"yes\",\n \"presence_type\": [\"children\"],\n \"estimated_people_count\": \"2-5\",\n \"visual_evidence\": \"Two children are visible standing side by side facing the waterfront.\"\n },\n \"nature_detection\": {\n \"nature_types\": [\"waterfront\", \"lawn\", \"woodland\"],\n \"dominant_nature_type\": \"waterfront\",\n \"visual_evidence\": \"A large body of water is visible beyond the lawn and trees, with a grassy area and scattered trees in the foreground.\"\n },\n \"activity_evidence\": {\n \"activity_types\": [\"watching\", \"landscape viewing\"],\n \"dominant_activity\": \"landscape viewing\",\n \"visual_evidence\": \"The children are standing still and looking out over the water and landscape.\"\n },\n \"cultural_experiential_meaning\": {\n \"meaning_types\": [\"aesthetic appreciation\", \"recreation\", \"relaxation\"],\n \"dominant_meaning\": \"aesthetic appreciation\",\n \"visual_evidence\": \"The scene suggests appreciation of the natural landscape and a recreational outdoor setting.\"\n },\n \"hni_overall_interpretation\": {\n \"is_hni_present\": \"yes\",\n \"hni_strength\": \"moderate\",\n \"reason\": \"Visible human presence interacting with a natural waterfront landscape, engaging in landscape viewing and appreciation.\"\n },\n \"confidence\": \"high\",\n \"uncertainty_notes\": \"None\"\n}", "parsed": { "image_level_summary": "Two children stand by a stone balustrade overlooking a waterfront with trees and lawn under a partly cloudy sky.", "human_presence": { "visible": "yes", "presence_type": [ "children" ], "estimated_people_count": "2-5", "visual_evidence": "Two children are visible standing side by side facing the waterfront." }, "nature_detection": { "nature_types": [ "waterfront", "lawn", "woodland" ], "dominant_nature_type": "waterfront", "visual_evidence": "A large body of water is visible beyond the lawn and trees, with a grassy area and scattered trees in the foreground." }, "activity_evidence": { "activity_types": [ "watching", "landscape viewing" ], "dominant_activity": "landscape viewing", "visual_evidence": "The children are standing still and looking out over the water and landscape." }, "cultural_experiential_meaning": { "meaning_types": [ "aesthetic appreciation", "recreation", "relaxation" ], "dominant_meaning": "aesthetic appreciation", "visual_evidence": "The scene suggests appreciation of the natural landscape and a recreational outdoor setting." }, "hni_overall_interpretation": { "is_hni_present": "yes", "hni_strength": "moderate", "reason": "Visible human presence interacting with a natural waterfront landscape, engaging in landscape viewing and appreciation." }, "confidence": "high", "uncertainty_notes": "None" }, "parse_error": null, "tokens_used": 1351 } } } }