Msajili mnunuzi mara moja wakati wa ununuzi. Kila skani ya lango ni mechi ya uso ya 1:N chini ya sekunde 1. Bure kwa kila skani baada ya usajili. Uthibitishaji 500 bila malipo kila mwezi.
Inaaminika na mashirika 2,000+ duniani kote.
Viwanja · Tamasha · Usafiri · Mikutano
Hakuna kutoa simu, hakuna wasiwasi wa betri, hakuna udanganyifu wa kuuza tena. Mnunuzi anajiandikisha mara moja wakati wa ununuzi kwa $0.25, na kila skani ya lango ya baadaye ni mechi ya bure ya Utafutaji wa Uso 1:N , chini ya sekunde 1 kwa kiwango cha tamasha. Mfumo huo huo hufanya kazi katika viwanja, sherehe, njia za usafiri, na mikutano.
Chagua ukaguzi unaotaka, ID, liveness, face match, vikwazo, anwani, umri, simu, barua pepe, maswali maalum. Ziburute kwenye mtiririko kwenye dashibodi, au tuma mtiririko huo huo kwenye API yetu. Panga masharti, fanya majaribio ya A/B, hakuna code inayohitajika.
$ curl -X POST https://verification.didit.me/v3/session/ \
-H "x-api-key: $DIDIT_API_KEY" \
-d '{
"workflow_id": "wf_event_enrol",
"vendor_data": "buyer-42",
"metadata": { "event_id": "evt_1234" }
}'{ "url": "verify.didit.me/..." }$ curl -X POST https://verification.didit.me/v3/face-search/ \
-H "x-api-key: $DIDIT_API_KEY" \
-F "image=@gate-frame.jpg" \
-F "metadata={'event_id':'evt_1234'}"{ "status": "Imeidhinishwa", "matches": […] }You are integrating Didit's biometric event-ticketing into <my_stack>. The face IS the ticket — no QR code, no barcode, no printout. Two phases:
1. Enrol the buyer ONCE at ticket purchase. A Sessions API call runs ID Verification + Passive Liveness and stores the buyer's portrait as the event-index template.
2. At every gate scan, call the standalone Face Search 1:N endpoint with the camera frame. Sub-1-second match against the event index. Verdict returned inline.
Pricing (public):
- Enrolment (one-time per buyer): $0.15 ID Verification + $0.10 Passive Liveness = $0.25
- Face Search 1:N at the gate: free per scan
- First 500 verifications free every month, forever
PRE-REQUISITES
- Production API key from https://business.didit.me (sandbox key in 60s, no card).
- Webhook endpoint with Hash-based Message Authentication Code (HMAC) SHA-256 verification using the X-Signature-V2 header.
HMAC-SHA256 verification MUST run against the raw body bytes (the raw payload as Didit sent it) BEFORE any JSON parsing — re-serialising the parsed body changes whitespace and key order, which invalidates the signature. - A workflow_id from the Workflow Builder. The workflow MUST contain OCR (ID Verification) + LIVENESS (and optionally Anti-Money Laundering (AML) if you sell age-restricted access).
- One Didit account = one face index. Tag every enrolment with metadata.event_id so the gate-side search filters by the right event.
STEP 1 — Enrol at ticket purchase
POST https://verification.didit.me/v3/session/
Headers:
x-api-key: <your api key>
Content-Type: application/json
Body:
{
"workflow_id": "<your enrolment workflow>",
"vendor_data": "<your buyer id — must be unique per buyer>",
"callback": "https://<your-app>/ticketing/enrol/callback",
"metadata": {
"order_id": "<your order reference>",
"event_id": "<the event the buyer is entering>",
"ticket_tier": "<GA | VIP | accessible | etc>"
}
}
Response: 201 Created with the hosted verification link in url. Redirect the buyer. After the buyer completes ID Verification + Passive Liveness, the portrait is stored as their face template and bound to vendor_data.
STEP 2 — Read the signed verdict on enrolment completion
Body (excerpted):
{
"session_id": "<uuid>",
"vendor_data": "<your buyer id>",
"status": "Approved",
"decision": {
"id_verifications": [
{ "node_id": "id_verification_1", "status": "Approved", "document_type": "Passport" }
],
"liveness_checks": [
{ "node_id": "liveness_1", "status": "Approved", "method": "PASSIVE", "score": 94 }
]
}
}
Verify X-Signature-V2 BEFORE trusting the body. On Approved, mark the order as enrolled and bind the order to vendor_data in your application database.
Session status enum (exact case): Approved | Declined | In Review | Resubmitted | Expired | Not Finished | Kyc Expired | Abandoned.
STEP 3 — At every gate scan, match the face against the event index
POST https://verification.didit.me/v3/face-search/
Headers:
x-api-key: <your api key>
Body (multipart/form-data):
image <the camera frame captured at the gate>
vendor_data <optional — restrict the search to a specific buyer>
metadata <optional JSON — e.g. { "event_id": "<...>" } to filter>
Response (excerpted):
{
"status": "Approved",
"matches": [
{
"vendor_data": "<buyer id>",
"similarity_score": 0.96
}
]
}
Match outcomes the gate logic should handle:
- Match found, ticket valid, no prior scan → grant entry, mark scanned
- Match found, already inside → apply re-entry policy
- Match found, ticket expired or canceled → route to staff
- Match found, ticket resold to another buyer → block, route to staff
- No match (similarity below threshold) → block, fall back to backup ticket flow
STEP 4 — Reusable across every event the same buyer attends
Because the buyer's face template is bound to vendor_data, the SAME enrolment works for every future event the same buyer holds a ticket for. No re-enrolment, no re-payment. Reusable Credential semantics applied to the venue use case.
CONSTRAINTS
- Base URL for /v3/* endpoints is verification.didit.me (NOT apx.didit.me).
- Feature enum is UPPERCASE: OCR, LIVENESS, FACE_MATCH, AML, IP_ANALYSIS.
- Method enum is UPPERCASE: PASSIVE, FLASHING, ACTIVE_3D.
- Auth header is x-api-key (lowercase, hyphenated).
- Webhook signature header is X-Signature-V2 (NOT X-Signature).
- Status casing matches exactly: Approved, Declined, In Review, Expired, Not Finished, Resubmitted, Kyc Expired, Abandoned.
- The face template is irreversible (a one-way hash). End users can request deletion via the standard data-subject-request path.
ACCESS CONTROL CONSIDERATIONS
- For minor-attended events, run AGE_ESTIMATION during enrolment to gate age-restricted areas. See /solutions/age-verification.
- For high-value VIP areas, gate behind a Biometric Authentication step-up. See /solutions/biometric-2fa.
- For accessibility lanes, surface the ticket-tier metadata on the verdict payload and route the gate UI accordingly.
DATA-PROTECTION NOTES
- The enrolment selfie produces an irreversible face template; raw images are deleted unless retention is explicitly enabled.
- The gate frame is processed in memory and not persisted by default.
- Display the privacy notice (didit.me/terms/verification-privacy-notice) to the buyer before the enrolment selfie. EU GDPR Article 13 requires it.
Read the docs:
- https://docs.didit.me/core-technology/face-search/overview
- https://docs.didit.me/sessions-api/create-session
- https://docs.didit.me/integration/webhooks
Start free at https://business.didit.me — sandbox key in 60 seconds, 500 verifications free every month, no credit card.
Kwa ajili ya kuunda, kujaribu, na watumiaji wako wa kwanza.
Moduli 25+, bei wazi. Punguzo la kiotomatiki kwa wingi.
Kwa kiasi kikubwa na programu zilizodhibitiwa.
Punguzo la wingi hutumika kiotomatiki kadri matumizi yanavyoongezeka — hakuna mazungumzo, hakuna simu ya mauzo.
Didit ni miundombinu ya utambulisho na udanganyifu, jukwaa ambalo tulitamani lingekuwepo tulipokuwa tukijenga bidhaa sisi wenyewe: wazi, rahisi kubadilika, na rafiki kwa waendelezaji, ili lifanye kazi kama sehemu halisi ya stack yako badala ya black box unayounganisha.
API moja inashughulikia kuthibitisha watu (KYC, know your customer), kuthibitisha biashara (KYB, know your business), kuchunguza pochi za crypto (KYT, know your transaction), na kufuatilia miamala kwa wakati halisi, kwenye stack iliyojengwa kuwa:
Nyayo chini: aina 14,000+ za nyaraka katika lugha 48+, vyanzo vya data 1,000+, na ishara za udanganyifu 200+ kwenye kila kipindi. Miundombinu ya Didit hujifunza kwa nguvu kutoka kila kipindi na inaboresha kila siku.
Tiketi ambapo tokeni ya kufikia ni uso wa mnunuzi, si Quick Response (QR) code, barcode, au chapisho la karatasi.
Mnunuzi anapiga selfie wakati wa ununuzi. Template ya uso inahifadhiwa dhidi ya agizo. Kwenye lango, kamera inachukua fremu moja na utafutaji wa uso wa 1:N unalinganisha na templates zilizosajiliwa za tukio. Uamuzi chini ya sekunde 1, pitia.
Matatizo matatu na tiketi zinazotegemea QR:
Mtiririko kamili kwa kawaida huchukua chini ya sekunde 30 `end-to-end`, chukua kitambulisho, piga picha ya hati, piga selfie, umemaliza. Hiyo ndiyo kasi zaidi sokoni. Watoa huduma wa zamani wa KYC kwa kawaida huchukua zaidi ya sekunde 90 kwa mtiririko huo huo.
Kwenye back end, Didit inarudisha matokeo kwa chini ya sekunde mbili kwa p99, ikipimwa kutoka wakati mtumiaji anamaliza selfie hadi wakati webhook yako inapoanza. Kunasa kwa simu kumerekebishwa kwa simu za polepole na mitandao ya polepole: progressive image compression, lazy software development kit load, na one-tap hand-off kutoka desktop kwenda simu kupitia QR code ikiwa mtumiaji anaanza kwenye web.
Daima tumia njia mbadala isiyo ya kibayometriki, PDF iliyochapishwa, QR code iliyotumwa kwa barua pepe, au lango lenye wafanyakazi. Sheria ya Ujasusi Bandia ya Umoja wa Ulaya na mifumo mingi ya kitaifa ya ulinzi wa watumiaji inahitaji njia mbadala.
Kwa vitendo, kumbi zinazotumia tiketi za kibayometriki zinaripoti 70-85% ya watu wanaokubali kwa sababu njia ya uso ni haraka zaidi. Njia isiyo ya kibayometriki inabaki wazi kwa wengine.
Kila kikao kinafikia mojawapo ya hali saba zilizo wazi, kwa hivyo code yako inajua nini cha kufanya kila wakati:
Approved, kila ukaguzi umefaulu. Mpeleke mtumiaji mbele.Declined, ukaguzi mmoja au zaidi umeshindwa. Unaweza kumruhusu mtumiaji kutuma tena hatua maalum iliyoshindwa (kwa mfano, piga tena selfie) bila kurudia mtiririko mzima.In Review, imewekwa alama kwa ukaguzi wa kufuata. Fungua kesi kwenye console, angalia kila ishara, amua kuidhinisha au kukataa.In Progress, mtumiaji yuko katikati ya mtiririko.Not Started, kiungo kimetumwa, mtumiaji bado hajafungua. Tuma ukumbusho ikiwa inakaa muda mrefu sana.Abandoned, mtumiaji alifungua kiungo lakini hakumaliza kwa wakati. Mshirikishe tena au muda umekwisha.Expired, kiungo cha kikao kimepitwa na wakati. Unda kikao kipya.Webhook iliyosainiwa inawaka kwenye kila mabadiliko ya hali, kwa hivyo database yako inabaki sawa kila wakati. Vikao vilivyoachwa na vilivyokataliwa ni bure.
Data ya uzalishaji huchakatwa na kuhifadhiwa katika Umoja wa Ulaya kwa chaguo-msingi, kwenye Amazon Web Services. Mikataba ya biashara inaweza kuomba maeneo mbadala kwa mamlaka ambazo wasimamizi wake wanahitaji.
Usimbaji fiche kila mahali. AES-256 ikiwa imetulia kwenye kila database, hifadhi ya object, na backup. Transport Layer Security 1.3 ikiwa safarini kwenye kila API call, webhook, na kipindi cha Business Console. Data ya kibayometriki imesimbwa kwa kutumia Customer Master Key tofauti.
Uhifadhi ni wako kudhibiti. Uhifadhi chaguomsingi ni usio na kikomo (unlimited) isipokuwa ukisanidi muda mfupi zaidi, kati ya siku 30 na miaka 10 kwa kila application, na unaweza kufuta kipindi chochote cha kibinafsi wakati wowote kutoka kwenye dashboard au API.
Vyeti: SOC 2 Type 1 & Type 2, ISO/IEC 27001:2022, iBeta Level 1 PAD, na uthibitisho wa umma kutoka Tesoro / SEPBLAC / CNMV ya Hispania kwamba uthibitishaji wa utambulisho wa mbali wa Didit ni salama zaidi kuliko kumthibitisha mtu ana kwa ana. Ripoti kamili inapatikana /security-compliance.
Didit inatii mahitaji kwa chaguo-msingi kwa wadhibiti muhimu kwa miundombinu ya utambulisho:
controller / processor split, Mkataba kamili wa Uchakataji Data ulichapishwa, mamlaka kuu ya usimamizi imetajwa (AEPD ya Uhispania).crypto on-ramps, exchanges, na custodians.age-gating na majukumu ya usalama wa watoto.Memo ya kina, kila cheti, kila barua ya mdhibiti: /security-compliance.
Njia tatu za kuunganisha, chagua inayofaa stack yako:
Dashibodi moja, bili moja, bei sawa ya kulipa-kwa-mafanikio kwa zote tatu. Mwongozo wa hatua kwa hatua kwenye docs.didit.me/integration/integration-prompt.
event indexes katika makumi ya maelfu ya nyuso zilizosajiliwa. Kasi ya kupita kwenye lango moja inakaa karibu viingilio 30 kwa dakika kwa kasi ya kutembea vizuri, inalinganishwa na QR scanner pipelines za haraka zaidi bila muda wa kutafuta. Platform uptime ni 99.99%, na operations playbook sawa na moduli nyingine zote za Didit.on-premise au hardware maalum?Kamera yoyote ya kawaida ya IP au tablet inafanya kazi, POST /v3/face-search/ inakubali fremu moja ya JPEG kupitia HTTPS. Kumbi nyingi hutumia entry kiosks zilizopo, ticket-scanner replacements, au tablet kwenye stendi. Hakuna Software Development Kit (SDK) maalum, hakuna drivers maalum, hakuna on-premise inference box.
Kwa edge-cached gates (usafiri, kumbi kubwa zisizo na muunganisho wa kuaminika), Didit inatoa on-premise face-search appliance, zungumza na mauzo.
Endesha AGE_ESTIMATION wakati wa usajili kwa maeneo yenye vizuizi vya umri (maeneo ya zaidi ya miaka 18 kwenye tamasha, baa za pombe kwenye uwanja). Uamuzi uliosainiwa hubeba age band kwa kila mnunuzi, na gate logic yako inaonyesha rangi sahihi ya wristband au njia isiyo na wristband.
Kwa watoto wadogo wanaohudhuria matukio ya familia, mzazi / mlezi anakamilisha usajili kwenye tiketi ya mtoto na kukubali template kwa niaba ya mtoto, Kanuni ya Jumla ya Ulinzi wa Data ya EU inahitaji idhini ya wazi ya mzazi chini ya umri wa miaka 16 katika nchi nyingi wanachama.
API moja kwa KYC, KYB, Ufuatiliaji wa Miamala, na Uchunguzi wa Wallet. Unganisha ndani ya dakika 5.