Plainsight
HomeOn-Demand LoginSupportRequest a Demo
Search…
Welcome
What's New
Getting Started
Start Here
Collect Data
Organize Datasets
Connect Datasources
Label
Define Your Labels
Labeling Your Images
Label Accelerators
Review Labels
Export Labels
Versioning
Train
Train Your Model with SmartML™
Train a Mobile Model
Test Your Model
Deploy Your Model
Predictions API
Google Pub/Sub
Walk-through: Training an Object Detection Model
Walk-through: Training a Classification Model
Walk-through: Training a Regression Model
Deploy
Pipelines
Account
Account Management
User Management
Billing
API Key Management
Private Cloud
GCP Private Cloud Users
Professional Services
Cisco Meraki Camera Guide
Powered By GitBook
Predictions API
This feature is being deprecated and will be replaced with Plainsight Pipelines static API.

Model Inferencing

Deployed models are available for inferencing using the Image Prediction API method, and the Video Prediction API method.
A generic object detection model is also available for inferencing. Read more about how to use this pre-trained model to return predictions on your data.

Authentication

Admins or Project Managers can generate an API Key in Settings to use for authentication. See the "Authorization" header below for details on using this key as a token.
post
https://<HOSTNAME>
/v1/models/:modelVersionId/predict
Image Prediction
Supported image types: JPG, PNG, GIF, BMP, TIFF, WEBP. File size limit is 10 MB.

Predict Image Examples

Send base64-encoded image as JSON:
curl -X POST 'https://<HOSTNAME>/v1/models/01F5PXD35EF2F8SBK27NM9ABC4/predict' \
--header 'Authorization: Bearer <APIKEY>' \
--header 'Content-Type: application/json' \
--data-raw '{ "image": "RmgBDRSUtAgpKO9LQAlsKoisdiVzVNME1wQ2VoaUh6cmVTek5UY3" }'
Send image as form data:
curl -X POST 'https://<HOSTNAME>/v1/models/01F5PXD35EF2F8SBK27NM9ABC4/predict' \
--header 'Authorization: Bearer <APIKEY>' \
--form '[email protected]/to/image.jpg'

Using the Generic Object Detection Model

Plainsight has a generic object detection model that can be used to AutoLabel your data and for inference. This model is pre-trained on the COCO dataset (a large dataset of common objects) and can return detections as bounding boxes or instance segmentation label types.
To use the general object detection inference model, specify one of the following model IDs in the modelVersionID path parameter of you request:
01ERDH3S3TZ34S863N47RWECA7 for detections labeled using bounding boxes (rectangles)
01ERDH3S45QPEG9YAXM5Q0RQ7W for detections labeled using instance segmentation (polygons)
For example, to run inference on the general object detection model and return detections as bounding boxes:
curl -X POST 'https://<HOSTNAME>/v1/models/01ERDH3S3TZ34S863N47RWECA7/predict' \
--header 'Authorization: Bearer <APIKEY>' \
--form '[email protected]/to/image.jpg'
post
https://<HOSTNAME>
/v1/models/:modelVersionId/batch-predict
Batch (Asynchronous) Video Prediction
get
https://<HOSTNAME>
/v1/models/:modelVersionId/batch-predict/:jobId/status
Job Status

Detections Output

Detections are output into a JSONL file. The URL to download this file is returned in the Job Status method upon successful completion of the job. Each line represents a frame of video.
  • time - the timestamp of the frame
  • frameIdx - index of frame within the video
  • detection object(s):
    • type - label type (rectangle, polygon, etc)
    • data - coordinates of the label
    • score - confidence score for this detection, ranging from 0.0-1.0
    • trackId - tracking ID used to track an object from frame to frame
Example output:
{"time": "00:00:00", "frameIdx": 0, "car": [{"type": "rectangle", "data": {"min": [0.7001143646240234, 0.7087980821397569], "max": [0.742496337890625, 0.7544862874348959]}, "children": {}, "score": 0.8047651648521423, "trackId": 0}, {"type": "rectangle", "data": {"min": [0.7988925170898438, 0.7319180636935764], "max": [0.8485193634033203, 0.7780952284071181]}, "children": {}, "score": 0.7955709099769592, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.0, 0.76564208984375], "max": [0.025728864669799803, 0.8191375732421875]}, "children": {}, "score": 0.7955175042152405, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.94644287109375, 0.7661393229166666], "max": [0.9950425720214844, 0.8066242811414931]}, "children": {}, "score": 0.6790698766708374, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.14806840896606446, 0.7507535129123264], "max": [0.22439971923828125, 0.8366963704427083]}, "children": {}, "score": 0.6356155872344971, "trackId": 4}, {"type": "rectangle", "data": {"min": [0.7422527313232422, 0.7078128390842013], "max": [0.7783774566650391, 0.7608119032118056]}, "children": {}, "score": 0.6314104795455933, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.48819801330566404, 0.7427418348524305], "max": [0.5575634384155274, 0.8126653374565972]}, "children": {}, "score": 0.5785809755325317, "trackId": 6}]}
{"time": "00:00:00.120000", "frameIdx": 3, "car": [{"type": "rectangle", "data": {"min": [0.7001200103759766, 0.7078899468315972], "max": [0.7417404937744141, 0.7540592447916666]}, "children": {}, "score": 0.8145115971565247, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.7980601501464843, 0.7328105333116319], "max": [0.8478513336181641, 0.7785378689236111]}, "children": {}, "score": 0.80238276720047, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.0, 0.7655868191189236], "max": [0.02540128231048584, 0.8191928439670139]}, "children": {}, "score": 0.7748746871948242, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.9465829467773438, 0.7661695014105903], "max": [0.9949024963378906, 0.8065941704644097]}, "children": {}, "score": 0.6508281230926514, "trackId": 0}, {"type": "rectangle", "data": {"min": [0.7423059844970703, 0.7077564832899306], "max": [0.7783242034912109, 0.7608683268229167]}, "children": {}, "score": 0.6322560906410217, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.489085693359375, 0.7427151150173611], "max": [0.5574259567260742, 0.8140271674262153]}, "children": {}, "score": 0.6002546548843384, "trackId": 6}, {"type": "rectangle", "data": {"min": [0.14681857109069824, 0.7657845052083333], "max": [0.21683483123779296, 0.8350164794921875]}, "children": {}, "score": 0.5051438212394714, "trackId": 4}]}
{"time": "00:00:00.200000", "frameIdx": 5, "car": [{"type": "rectangle", "data": {"min": [0.7979566192626953, 0.7319490559895834], "max": [0.8479548645019531, 0.7780642361111111]}, "children": {}, "score": 0.8713392019271851, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.7000874328613281, 0.7078708224826389], "max": [0.7417730712890624, 0.7540784369574652]}, "children": {}, "score": 0.8168344497680664, "trackId": 4}, {"type": "rectangle", "data": {"min": [0.0, 0.7655344645182292], "max": [0.025421178340911864, 0.8192451985677084]}, "children": {}, "score": 0.7828996181488037, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.9465740966796875, 0.766183336046007], "max": [0.9949113464355469, 0.8065802680121528]}, "children": {}, "score": 0.6592091917991638, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.7421274566650391, 0.70782958984375], "max": [0.7785027313232422, 0.7607952202690972]}, "children": {}, "score": 0.6189853549003601, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.48903560638427734, 0.7427429877387153], "max": [0.5574760437011719, 0.8139992947048611]}, "children": {}, "score": 0.5979089736938477, "trackId": 6}]}
{"time": "00:00:00.320000", "frameIdx": 8, "car": [{"type": "rectangle", "data": {"min": [0.7988416290283203, 0.7318497043185764], "max": [0.8485702514648438, 0.778163587782118]}, "children": {}, "score": 0.8568191528320312, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.6996240997314453, 0.7078053114149305], "max": [0.7422364044189453, 0.7541439480251736]}, "children": {}, "score": 0.8298211693763733, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.0, 0.765592041015625], "max": [0.025442245006561278, 0.8191876220703125]}, "children": {}, "score": 0.7620850205421448, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.9466223907470703, 0.7661448160807292], "max": [0.9948629760742187, 0.8066188557942708]}, "children": {}, "score": 0.690345823764801, "trackId": 4}, {"type": "rectangle", "data": {"min": [0.7422945404052734, 0.7079161919487847], "max": [0.7783356475830078, 0.7607086181640625]}, "children": {}, "score": 0.5972740054130554, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.14873595237731935, 0.750741441514757], "max": [0.22373218536376954, 0.8367084418402778]}, "children": {}, "score": 0.5879378318786621, "trackId": 7}, {"type": "rectangle", "data": {"min": [0.4890005111694336, 0.7425733100043402], "max": [0.5575111389160157, 0.8128338623046875]}, "children": {}, "score": 0.5386565923690796, "trackId": 6}]}
{"time": "00:00:00.400000", "frameIdx": 10, "car": [{"type": "rectangle", "data": {"min": [0.6995636749267579, 0.7082058376736111], "max": [0.7422968292236328, 0.753743421766493]}, "children": {}, "score": 0.8554792404174805, "trackId": 4}, {"type": "rectangle", "data": {"min": [0.7996922302246093, 0.7325918240017361], "max": [0.8492200469970703, 0.778756578233507]}, "children": {}, "score": 0.806079089641571, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.0, 0.7671257866753473], "max": [0.025510644912719725, 0.8189889865451389]}, "children": {}, "score": 0.6845098733901978, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.9361000061035156, 0.7669605170355903], "max": [0.9963831329345703, 0.8058031548394097]}, "children": {}, "score": 0.6576622128486633, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.7422310638427735, 0.7085048421223958], "max": [0.7783990478515626, 0.7601199001736111]}, "children": {}, "score": 0.6540669202804565, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.1461125946044922, 0.795661146375868], "max": [0.1805940818786621, 0.8291719563802084]}, "children": {}, "score": 0.6369971632957458, "trackId": 8}, {"type": "rectangle", "data": {"min": [0.14645214080810548, 0.7578844536675348], "max": [0.22301523208618165, 0.8335708279079861]}, "children": {}, "score": 0.5972306132316589, "trackId": 7}, {"type": "rectangle", "data": {"min": [0.09434420585632325, 0.7698390706380208], "max": [0.12827395439147948, 0.8162757025824653]}, "children": {}, "score": 0.5838125944137573, "trackId": 9}, {"type": "rectangle", "data": {"min": [0.4893690872192383, 0.7425895182291666], "max": [0.5571425628662109, 0.8141527642144097]}, "children": {}, "score": 0.5766813158988953, "trackId": 6}, {"type": "rectangle", "data": {"min": [0.5697303009033203, 0.7300347900390625], "max": [0.6215675354003907, 0.7652922905815972]}, "children": {}, "score": 0.5320419073104858, "trackId": 10}]}
{"time": "00:00:00.520000", "frameIdx": 13, "car": [{"type": "rectangle", "data": {"min": [0.1442272663116455, 0.7567987738715278], "max": [0.2228019905090332, 0.8333213297526042]}, "children": {}, "score": 0.8348209261894226, "trackId": 7}, {"type": "rectangle", "data": {"min": [0.7987076568603516, 0.7309903293185764], "max": [0.8487042236328125, 0.7776878526475695]}, "children": {}, "score": 0.8244227766990662, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.0, 0.7662373860677083], "max": [0.02554854154586792, 0.8198773871527778]}, "children": {}, "score": 0.7334166169166565, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.7007211303710937, 0.7085287814670139], "max": [0.7418894958496094, 0.7547555881076389]}, "children": {}, "score": 0.7187542915344238, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.7418098449707031, 0.7061136881510417], "max": [0.7773199462890625, 0.7598409016927083]}, "children": {}, "score": 0.6801111102104187, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.9350718688964844, 0.7669188096788194], "max": [0.9959108734130859, 0.8058448621961806]}, "children": {}, "score": 0.6667593121528625, "trackId": 4}, {"type": "rectangle", "data": {"min": [0.5675093841552734, 0.7281532118055556], "max": [0.6297899627685547, 0.7671738688151042]}, "children": {}, "score": 0.6150084733963013, "trackId": 10}, {"type": "rectangle", "data": {"min": [0.09586554527282715, 0.76857421875], "max": [0.12937828063964843, 0.8162054443359374]}, "children": {}, "score": 0.6082856059074402, "trackId": 9}, {"type": "rectangle", "data": {"min": [0.14507335662841797, 0.7785641818576389], "max": [0.1825710678100586, 0.8329177517361112]}, "children": {}, "score": 0.5406203866004944, "trackId": 8}]}
{"time": "00:00:00.640000", "frameIdx": 16, "car": [{"type": "rectangle", "data": {"min": [0.14417657852172852, 0.7569097222222222], "max": [0.2226651382446289, 0.8332103814019097]}, "children": {}, "score": 0.8460432887077332, "trackId": 8}, {"type": "rectangle", "data": {"min": [0.7004826354980469, 0.7091966417100695], "max": [0.7421280670166016, 0.7540877278645833]}, "children": {}, "score": 0.806346595287323, "trackId": 7}, {"type": "rectangle", "data": {"min": [0.7997734069824218, 0.7327389865451389], "max": [0.8476384735107422, 0.775939195421007]}, "children": {}, "score": 0.7701347470283508, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.0, 0.7660401746961806], "max": [0.025666043758392335, 0.8200745985243055]}, "children": {}, "score": 0.758480429649353, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.7415583038330078, 0.7069877794053819], "max": [0.7768212890625, 0.7603018527560764]}, "children": {}, "score": 0.6612362861633301, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.5687237167358399, 0.7294388156467014], "max": [0.6225741195678711, 0.7658882649739583]}, "children": {}, "score": 0.6109995245933533, "trackId": 4}, {"type": "rectangle", "data": {"min": [0.09434274673461913, 0.7701981608072916], "max": [0.1290256118774414, 0.8159166124131945]}, "children": {}, "score": 0.5956371426582336, "trackId": 9}, {"type": "rectangle", "data": {"min": [0.9266272735595703, 0.7696741400824653], "max": [0.9953533172607422, 0.8030894639756945]}, "children": {}, "score": 0.5858432054519653, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.4889754104614258, 0.7424461534288195], "max": [0.5567860412597656, 0.8129610188802083]}, "children": {}, "score": 0.5347800850868225, "trackId": 6}, {"type": "rectangle", "data": {"min": [0.14479297637939453, 0.7775484212239583], "max": [0.18285144805908204, 0.8325984022352431]}, "children": {}, "score": 0.5075516700744629, "trackId": 10}]}
{"time": "00:00:00.720000", "frameIdx": 18, "car": [{"type": "rectangle", "data": {"min": [0.7993070220947266, 0.7322889539930556], "max": [0.8511056518554687, 0.7777243381076389]}, "children": {}, "score": 0.8554795980453491, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.7002946472167969, 0.710480482313368], "max": [0.7423159790039062, 0.7541389295789931]}, "children": {}, "score": 0.7996271252632141, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.0, 0.7663991292317708], "max": [0.025371847152709962, 0.8197156439887153]}, "children": {}, "score": 0.7976168394088745, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.5691310882568359, 0.7296035427517361], "max": [0.622166748046875, 0.7657234700520833]}, "children": {}, "score": 0.6927707195281982, "trackId": 9}, {"type": "rectangle", "data": {"min": [0.7416486358642578, 0.7074123806423611], "max": [0.77673095703125, 0.7598773193359375]}, "children": {}, "score": 0.6590728163719177, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.15913079261779786, 0.7441860622829861], "max": [0.2365931510925293, 0.8312477620442709]}, "children": {}, "score": 0.6562774777412415, "trackId": 8}, {"type": "rectangle", "data": {"min": [0.14653841018676758, 0.7940071614583334], "max": [0.18054336547851563, 0.8294908311631944]}, "children": {}, "score": 0.617221474647522, "trackId": 10}, {"type": "rectangle", "data": {"min": [0.48919025421142576, 0.7426589626736111], "max": [0.5573213958740234, 0.8127482096354167]}, "children": {}, "score": 0.5728475451469421, "trackId": 6}, {"type": "rectangle", "data": {"min": [0.09469179153442382, 0.7692333306206597], "max": [0.1284890079498291, 0.8155463324652777]}, "children": {}, "score": 0.5208727717399597, "trackId": 4}]}
{"time": "00:00:00.800000", "frameIdx": 20, "car": [{"type": "rectangle", "data": {"min": [0.7992606353759766, 0.7318089463975694], "max": [0.8496516418457031, 0.778204345703125]}, "children": {}, "score": 0.875287652015686, "trackId": 1}, {"type": "rectangle", "data": {"min": [0.00018331080675125123, 0.764676038953993], "max": [0.02326004981994629, 0.8187685139973958]}, "children": {}, "score": 0.8330589532852173, "trackId": 2}, {"type": "rectangle", "data": {"min": [0.7003234100341796, 0.7106207275390625], "max": [0.7422872924804688, 0.7539986843532986]}, "children": {}, "score": 0.8263353705406189, "trackId": 7}, {"type": "rectangle", "data": {"min": [0.7415248870849609, 0.7084487575954861], "max": [0.7768547821044922, 0.7601760525173611]}, "children": {}, "score": 0.7250340580940247, "trackId": 5}, {"type": "rectangle", "data": {"min": [0.5678661727905273, 0.7281162516276042], "max": [0.6294331741333008, 0.7672108289930556]}, "children": {}, "score": 0.7118926644325256, "trackId": 9}, {"type": "rectangle", "data": {"min": [0.14996875762939454, 0.7464505343967014], "max": [0.23318952560424805, 0.83432373046875]}, "children": {}, "score": 0.672606348991394, "trackId": 8}, {"type": "rectangle", "data": {"min": [0.48922512054443357, 0.7422311062282986], "max": [0.5580366897583008, 0.8145111762152778]}, "children": {}, "score": 0.6424439549446106, "trackId": 6}, {"type": "rectangle", "data": {"min": [0.14664278984069823, 0.7816626654730903], "max": [0.18137672424316406, 0.8324894883897569]}, "children": {}, "score": 0.5970743894577026, "trackId": 10}, {"type": "rectangle", "data": {"min": [0.9294758605957031, 0.7699845377604166], "max": [0.99550537109375, 0.8027790662977431]}, "children": {}, "score": 0.5623807311058044, "trackId": 3}, {"type": "rectangle", "data": {"min": [0.09595375061035157, 0.7683754475911458], "max": [0.12910252571105957, 0.8164042154947917]}, "children": {}, "score": 0.5287332534790039, "trackId": 4}]}
Train - Previous
Deploy Your Model
Next - Train
Google Pub/Sub
Last modified 1mo ago
Copy link
On this page
Model Inferencing
Authentication
post
Image Prediction
Predict Image Examples
Using the Generic Object Detection Model
post
Batch (Asynchronous) Video Prediction
get
Job Status
Detections Output