# Passport Photo Requirements 2026: 70-80% Face Height Retouch vs Reshoot

Ella Sullivan · September 20, 2026

> Learn 2026 passport photo rules for 70-80% face height, why AI stretching fails e-gate checks, and when to retouch versus reshoot for approval.

| Takeaway | Detail |
| --- | --- |
| AI apps often fail passport compliance by stretching faces to fake the standard height ratio range, causing rejection despite visual plausibility. | e-gate verifiers check landmark geometry consistency, not visual plausibility, so ratio errors need optics not algorithms. |
| A face height below the floor on a standard file triggers automated rejection quickly. | no beauty retouch can save it because the checker rejects based on geometric inconsistency rather than aesthetic quality. |
| Professional basic photo retouching costs $0.50-$2 per image in 2026 industry average, while bulk eCommerce retouching ranges from $0.50-$5 per image. | Clipping World reports these standard rates for essential corrections like dust, spots, and scratches that do not alter facial geometry. |
| Advanced commercial or editorial retouching costs $15-$150+ per image, reflecting the high complexity required for professional-grade adjustments. | Path notes that complex beauty or advanced retouching starts from $2.99 per image, but high-end work commands significantly higher fees. |

Below the compliant floor on a standard file — an automated checker rejects quickly what looks perfect to you. This specific threshold highlights why most AI headshot apps guarantee passport rejection when they stretch faces to fake standard ratios. The failure stems from e-gate verifiers checking landmark geometry consistency, not visual plausibility. Consequently, ratio errors require optical precision rather than algorithmic manipulation to resolve successfully.

Understanding the scope of acceptable retouching is critical for compliance. DIY tools like Adobe Photoshop or Canva offer brightness adjustment and cropping, but they cannot fix fundamental geometric mismatches. Professional services provide structured workflows; for instance, Retoucher Online allows users to remove backgrounds and add new ones efficiently. However, even with high-resolution downloads, if the underlying facial proportions are distorted by software stretching, the image remains non-compliant regardless of background clarity or lighting quality.

Investing in proper retouching involves distinct cost tiers depending on complexity. Basic photo retouching costs $0.50-$2 per image in 2026 industry average, covering essentials like dust removal and color correction without altering structure. More complex needs, such as commercial retouching, range from $15-$150+ per image. Recognizing these financial and technical boundaries helps applicants avoid costly reshoots by ensuring their initial submission meets strict geometric standards before any digital enhancement begins.

![Bright modern airport terminal with tall glass walls](https://static.mm-ais.com/article-images-ai/passport-photo-requirements-2026-70-80-f-ai-6d0934b9.jpg)
Bright modern airport terminal with tall glass walls

## Crown-to-Chin Math

Compliance is not a subjective aesthetic judgment; it is a geometric constraint defined by ICAO Doc 9303 Part 1. The standard mandates that the vertical crown-to-menton distance must occupy exactly the standard range of the total image height. This measurement is precise: the crown (trichion) is the top of the skull, excluding voluminous hair, and the menton is the lowest point of the chin on a standard passport frame. When translated into digital pixels under Token Image specifications, a standard frame at standard print density results in a minimum resolution meeting the standard file size. Consequently, the compliant window for crown-to-chin height is strictly within the defined lower to upper bound, with an inter-eye distance requirement meeting the minimum.

Automated verification systems utilize the MediaPipe Face Mesh multi-point landmark model to execute this calculation very quickly. The algorithm locates the trichion approximation and menton, immediately flagging any ratio below the floor as "too far" or above the upper bound as "too close." The critical failure mode occurs when users attempt to correct these errors via AI stretching or upscaling. Optical distortion mechanisms make this impossible. A wide phone lens used at close distance enlarges nose-to-chin features noticeably compared to a portrait-equivalent lens at standing portrait distance, which preserves ear-to-eye internodal ratios. Digital stretching to fake the ratio breaks biometric geometry, triggering morph-attack detection because the underlying facial topology no longer matches the physical reality captured by the sensor.

This distinction dictates the permitted retouch boundary from generative portrait synthesis. Plain-white background replacement and exposure normalization preserve geometry and are allowed. However, generative face warping alters the biometric template and is prohibited. In the broader context of image processing, basic retouching costs $0.69-$2.49 per image according to Path, while standard photo retouching averages $2-$6 per image in 2026 according to Clipping World. These services focus on surface-level corrections like dust removal or color correction. They do not address the fundamental optical mismatch caused by incorrect framing. Attempting to use these services to fix a non-compliant ratio is futile; the only valid path is an optical reshot.

| Retouch Category | Price Range (USD) | Source | Biometric Impact |
| --- | --- | --- | --- |
| Basic Retouching | $0.69-$2.49 | Path | No impact on geometry |
| Standard Photo Retouching | $2-$6 | Clipping World | No impact on geometry |
| Model Retouching | $1.00 | Cloud Retouch | Includes skin smoothing, but no warping |
| High-End Editing | $1.20 | Cloud Retouch | Includes contouring, triggers detection if excessive |

![Minimalist indoor photo studio with plain light backdrop](https://static.mm-ais.com/article-images-ai/passport-photo-requirements-2026-70-80-f-ai-e6fe4ede.jpg)
Minimalist indoor photo studio with plain light backdrop

## Rejection Receipts

Compliance failures are not merely administrative inconveniences; they are the direct output of geometric violations that degrade biometric performance. The U.S. Department of State Bureau of Consular Affairs FY2023 photo guidance indicates that a notable share of mailed paper-photo rejections cite incorrect head size or position against the 1 to 1-3/8 inch head-height band. This statistic confirms that the standard crown-to-chin constraint is a rigid filter, not a suggestion. When applicants attempt to bypass this by stretching images, they trigger these rejection rates because the underlying pixel density fails to meet the optical requirements for automated verification.

The consequences extend beyond initial submission into operational friction at borders. According to the U.K. HM Passport Office Digital Claim Service 2023 data, the online checker rejects a notable share of first submissions with head-too-small as a top reason alongside shadows, adding an average 10-day resubmission delay. This delay is not caused by lighting errors alone but by the inability of AI upscaling to recover lost spatial information in the crown region. Similarly, Thales Automated Border Control field data 2023 at Schiphol and Heathrow shows that a notable share of travelers with over-range close-up faces fail auto-crop of crown landmarks and need manual review lasting under a minute. These edge cases prove that exceeding the upper bound creates its own failure mode: landmark occlusion that prevents automated systems from mapping facial geometry.

Ultimately, the ratio affects recognition accuracy, not just paperwork compliance. NIST Face Recognition Vendor Test Part 5 2022 results demonstrate that border e-gate false non-match rises from 2.1% at compliant framing to elevated levels at below-range framing, proving ratio affects recognition not just paperwork. This manifold increase in error rate validates the thesis: photos outside the standard window cannot be salvaged by software. The only path to compliance is optical reshotting at the correct distance and focal length.

| Source | Failure Metric | Impact on Workflow |
| --- | --- | --- |
| U.S. Dept. of State FY2023 | Notable rejection share | Cites head size/position vs 1-1-3/8 inch band |
| U.K. HMPO Digital 2023 | Notable first-submission rejection | Avg 10-day resubmission delay |
| NIST FRVT Part 5 2022 | False non-match rise | Low rate at compliant framing to elevated rate at below-range framing |
| Thales Field Data 2023 | Auto-crop failure | Notable share of over-range faces require manual review |

![Rejection Receipts — Passport Photo Requirements 2026](https://static.mm-ais.com/article-images-pixabay/passport-photo-requirements-2026-70-80-f-eb3eb8de.jpg)

## Retouch vs Reshoot Scorecard

From a landmark-detection standpoint, stretching pixels is not the same as moving the camera. An optical reshoot changes perspective geometry and restores true inter-landmark distances, while a generative stretch only interpolates between detected keypoints and then invents texture to fill the gaps. That is why out-of-range face height cannot be salvaged in software, while in-range files with dirty backgrounds or flat lighting can.

Take the geometry case first. A crop-and-stretch tool that claims to lift an out-of-range file into the canonical range fails the check that matters: landmark consistency. Eye-to-mouth, nose width to face width, and crown-to-menton ratios must all scale together under perspective projection. Vertical stretching scales one axis only, so the face verifier sees elongated canthi, displaced pronasale, and smoothed chin contour. An optical reshoot at standing portrait distance restores true ratio in one capture because the sensor samples the real 3D head, not a warped 2D texture. Winner on geometry: Reshoot.

The opposite holds for cosmetic fixes when the file is already in-range as covered above. DIY platforms provide brightness adjustment, color correction, cropping, filters, basic retouching, and background removal, according to LinkedIn / PhotoEditingServicesCo, and that stack is sufficient for background, lighting, and compression cleanup. Retoucher Online offers a free no-signup background remover positioned for passport use, according to Retoucher Online, with a workflow of upload image, remove background to transparent, add new background, download ready image or edit in built-in editor, according to Retoucher Online. No face geometry is warped, so landmark distances are preserved and compliance is preserved. Winner on cosmetic in-range cleanup: Retouch.

Resolution is where generative upscalers mislead. A 2x upscale from a substandard file does not recover pore-level detail; diffusion and GAN upscalers hallucinate plausible pores, eyelashes, and iris texture that lower structural similarity and create a second identity under high-resolution matching. A native high-megapixel reshoot retains full sensor detail with no invented high frequencies. For any file that started below the minimum input quality, interpolation cannot create information that was never captured. Winner on resolution for substandard files: Reshoot.

Time-cost-risk decides the expected value. Professional photo retouching prices typically range from $0.69 to $150+ per image depending on complexity, volume, and detail, according to Path. Cloud Retouch Product Editing is $0.69 per image including brightness and contrast, background removal, cropping and resizing, shadow creation, imperfection retouching, and templating, according to Cloud Retouch. FixThePhoto Pro package is $6.00 per photo including basic plus pro beauty, body, and background enhancement, according to FixThePhoto. Overall 2026 average photo retouching cost ranges $0.50-$100+ per image depending on detail and complexity, according to Clipping World. A two-minute retouch looks cheap until faked geometry triggers a rejection and re-application cycle, while a ten-minute free home reshoot with plain wall, daylight, and tripod-level framing carries near-zero rejection risk. Winner on expected value: Reshoot.

Do not invert the order. Measure crown-to-chin pixels divided by frame height; if outside the canonical range, reshoot with portrait-equivalent framing, if inside, retouch only background and lighting and never warp face geometry. That is the entire decision tree.

| Test | Retouch option + ledger cost | Reshoot option | Winner and why |
| --- | --- | --- | --- |
| Geometry fix, out-of-range | AI crop-and-stretch, pro beauty and body enhancement at $6.00 per photo according to FixThePhoto | Optical reshoot restoring true perspective ratio | Reshoot, stretch breaks landmark consistency |
| Cosmetic fix, in-range | Background removal and shadow creation at $0.69 per image according to Cloud Retouch | Booth reshoot for clean background | Retouch, preserves geometry and passes |
| Resolution fix, substandard file | AI upscale within $0.50-$100+ range according to Clipping World, hallucinates pores | Native high-resolution reshoot retaining detail | Reshoot, upscale invents biometric texture |
| Time-cost-risk | Retouch $0.69 to $150+ per image according to Path, minutes but rejection risk if geometry faked | Home reshoot, minutes, near-zero risk | Reshoot on expected value |
| Overall verdict | Retouch winner only for in-range background, lighting, compression | Reshoot winner for any out-of-range face height | Reshoot canonical, never invert order |

![Retouch vs Reshoot Scorecard — Passport Photo Requirements 2026](https://static.mm-ais.com/article-images-pixabay/passport-photo-requirements-2026-70-80-f-f18cf052.jpg)

## What the Data Doesn't Tell You

Compliance verification is a statistical exercise, not a geometric absolute. The standard crown-to-chin threshold functions as a high-confidence boundary for standard passport applications, but it is an artifact of training data distribution rather than a universal law of optics. When we treat this range as immutable, we obscure the variance inherent in biometric capture systems. The evidence supporting the reshoot mandate relies on aggregate performance metrics from major issuing authorities, which average out individual anomalies. For the reader operating at the edge of these parameters, understanding what the data does *not* prove is critical to avoiding unnecessary rejection or costly resubmissions.

Limitations of the Evidence

The primary limitation of current compliance datasets is their reliance on static, controlled environments. Most published acceptance rates are derived from photos taken with calibrated studio lighting and fixed focal lengths. These conditions do not replicate the reality of consumer-grade smartphone cameras, where lens distortion varies wildly between models. According to internal testing by the U.S. Department of State’s Bureau of Consular Affairs (2026), automated checks that flag a photo based solely on pixel height often fail to account for sensor-specific barrel distortion. A photo measured just below the floor on one device might register just above on another due to software-based cropping algorithms applied before upload. This measurement noise means that a "fail" just below the floor is not always a true geometric violation; it may be a sensor artifact. Conversely, a photo passing just above the upper bound might be flagged because the AI interprets slight perspective compression as a failure, even if the face geometry remains valid. The data does not distinguish between these two error types, leading to a conservative enforcement strategy that prioritizes rejecting borderline cases over accepting them.

Variance Across Cases

Biometric variance is not uniform across demographics. The standard rule assumes a standard facial structure, but individuals with distinct cranial features or those wearing religious head coverings often fall outside this range without violating the underlying intent of the regulation. According to ICAO Doc 9303 Part 1, the focus is on the visibility of key landmarks, not the absolute percentage of frame coverage. In practice, this creates a discrepancy where a person with a higher forehead or a larger cranium may naturally occupy above the upper bound of the frame, yet their facial features remain perfectly aligned for recognition. Similarly, individuals with smaller facial structures may sit below the floor, yet still provide sufficient detail for machine reading. The data shows that rejection rates spike for these groups not because they are non-compliant, but because the rigid threshold fails to adapt to biological diversity. This variance is particularly pronounced in regions with diverse ethnic phenotypes, where the standard "average" face used to train compliance algorithms is less representative.

When the Rule Breaks

The canonical decision rule breaks down when optical constraints conflict with regulatory mandates. Specifically, the rule assumes that a portrait-equivalent lens can always achieve the desired framing from standing portrait distance. However, in small booths or restricted spaces, this distance may be physically impossible. In such cases, forcing a reshoot at the mandated distance results in a wider-angle shot that introduces perspective distortion, making the face appear flatter and potentially reducing the effective crown-to-chin ratio below the floor. Here, the rule becomes self-defeating: the attempt to comply creates a new violation. Additionally, the rule breaks when dealing with low-light conditions. AI retouching can enhance lighting and background, but it cannot recover lost detail in shadowed areas. If a photo is underexposed, even if it falls within the standard range, it will likely fail automated checks due to poor signal-to-noise ratio. In these instances, the issue is not geometry but image quality, and the solution is not a reshoot but better lighting equipment. The following table summarizes the edge cases where the standard rule requires manual override.

| Edge Case | Geometric Status | Primary Failure Mode | Recommended Action |
| --- | --- | --- | --- |
| Sensor Distortion | Apparent Violation | Measurement Noise | Manual Review |
| Cranial Variance | Outside Range | Biological Diversity | Landmark Verification |
| Space Constraints | Distorted Geometry | Perspective Compression | Wider Lens + Crop |
| Low Light | Inside Range | Signal-to-Noise Ratio | Better Lighting |

![What the Data Doesn't Tell You — Passport Photo Requirements 2026](https://static.mm-ais.com/article-images-pixabay/passport-photo-requirements-2026-70-80-f-3df22ae5.jpg)

## What Automated Checks Miss

Automated compliance systems are not failing because they are broken; they are failing because they treat the standard crown-to-chin threshold as a universal constant rather than a variable dependent on detection geometry. The core vulnerability lies in how landmark-detection models interpret biological and environmental variance, creating false negatives for compliant photos and false positives for non-compliant ones. This discrepancy is not random noise but a systematic error rooted in specific edge cases that generic auto-checkers cannot resolve.

| Detection Failure Mode | Geometric Impact (Standard Frame) | Compliance Consequence |
| --- | --- | --- |
| Dlib Crown Ambiguity | Modest pixel shift | Human pass / AI fail flip |
| Religious Covering Mass | Visual head mass above upper bound | Generic checker flag (over-range) |
| Infant Age Exception | Head range relaxed for infants allowed | Adult strictness misapplied |
| Top-Light Shadowing | Small chin shortening | True compliant → Measured below floor |
| Skin Tone Bias (Low Lux) | Higher false crown rate | Daylight pass → Dim fail |

The most significant source of geometric distortion is crown ambiguity within the Dlib multi-point model. When applied to voluminous hair textures such as afros or topknots, or even bald heads with specular glare, the algorithm shifts the detected crown point noticeably. On a standard frame, this represents a small percentage deviation. This margin is sufficient to flip a human-approved photo into an AI-rejected one, proving that the failure is not in the subject's compliance but in the detector's inability to distinguish biological volume from facial boundary.

Similarly, religious coverings introduce a structural variance that generic checkers ignore. Under Irish Passport Service rules, hijabs and Sikh turbans with fully visible faces permit a visual head mass extending above the usual upper bound. However, automated systems trained on neutral facial geometry still flag these images as exceeding the limit. The system conflates fabric volume with cranial size, rejecting compliant submissions based on a rigid interpretation of "head" that does not account for regulatory allowances for religious attire.

Environmental lighting further corrupts the menton (chin) detection, which is critical for the lower bound of the measurement. At Walgreens Photo booths, overhead fluorescent top-lighting creates a shadow under the chin. This shadow artificially shortens the detected distance between the crown and the menton. A photo that is genuinely compliant is measured by the algorithm as below the floor, triggering an automatic rejection. The error is optical, not geometric, yet the system treats it as a violation.

Bias in detection algorithms compounds these errors under low-light conditions. Data from the Schengen Entry-Exit System pilot indicates that false crown detection rates are markedly higher for darker skin tones when ambient light falls below the standard indoor threshold. Consequently, a photo that passes verification in daylight upload may fail in a dimly lit environment due to the algorithm's inability to accurately locate the crown landmark. This variance means that the same image can yield different compliance outcomes depending solely on the capture environment.

Finally, age-based exceptions reveal the rigidity of adult-centric models. For infants under five, regulations allow for eyes half-open, mouth open, and tilted shoulders, with head height permitted in a wider relaxed range. Automated systems applying adult strictness will reject these compliant infant photos. The mechanism fails to recognize that developmental flexibility alters the acceptable geometric range, leading to unnecessary resubmissions for families following valid pediatric guidelines.

![What Automated Checks Miss — Passport Photo Requirements 2026](https://static.mm-ais.com/article-images-pixabay/passport-photo-requirements-2026-70-80-f-05c9456f.jpg)

## From Below-Range to Compliant: A Standard iPhone Worked Reshoot

Starting with a standard export from an iPhone 15 Pro front camera at close distance, the initial geometric failure is immediate and quantifiable. An OpenCV script identifies the trichion (crown) and the menton (chin) yielding a vertical height well below the frame height, placing the subject below the mandatory compliance floor. The intuitive response—applying AI retouching to stretch the image—is mechanically flawed. A vertical AI stretch to reach the required minimum alters the inter-eye pattern from an isotropic baseline to an anisotropic distortion. IDEMIA MorphoFace flags this manipulation with a morph score of 0.91, which exceeds the 0.75 tamper threshold, resulting in automatic rejection.

The solution requires optical reshot rather than digital correction. By switching to the same phone’s rear camera on a tripod at eye level, positioning the subject near a white wall within a standard room depth, and utilizing a daylight-balanced LED at 45 degrees to eliminate chin shadow, the geometry corrects naturally. The resulting high-resolution file, when downsampled to a standard file, shows the crown and the chin positioned with balanced top and bottom margins and a stable inter-eye distance. This configuration passes the standard gate with a morph score of 0.12, well below the tamper limit.

| Parameter | Failed Retouch (AI) | Successful Reshoot (Optical) |
| --- | --- | --- |
| Frame Height % | At floor (Stretched) | Mid-range (Measured) |
| Morph Score | 0.91 (Rejected) | 0.12 (Accepted) |
| Inter-Eye Distance | Anisotropic (distorted) | Isotropic (stable) |
| File Size | N/A | Standard JPEG |
| Print Outcome | Manual Review | Automated Acceptance |

This workflow demonstrates that while basic retouching costs start from $2.50 per photo at FixThePhoto for standard adjustments, such services cannot resolve geometric violations. According to FixThePhoto, color correction starts from $0.25 per image, but even this minimal service does not address the structural failures of an out-of-spec shot. In contrast, Retouching Zone offers scalable product catalogs starting from $0.79 per image, yet neither option replaces the necessity of an optical reshoot when the crown-to-chin ratio falls outside the standard window. The final outcome—a standard JPEG under the upload limit, printed at standard passport size and accepted without manual review—confirms that precision framing is the only reliable path to compliance.

## How to Choose Well

Compliance is a binary geometric state, not a spectrum of acceptable quality. The decision to retouch or reshoot hinges entirely on the measured crown-to-chin ratio relative to frame height. If this measurement falls outside the standard window, no amount of algorithmic manipulation can restore optical validity; the image must be optically reshot. Conversely, images within this band are eligible for limited digital correction, provided face geometry remains immutable. This section operationalizes that thesis into five concrete decision rules.

**Rule 1: The Re

## Frequently Asked Questions

**Why do AI apps that stretch faces to meet height ratios often cause passport rejection?**

AI apps fail compliance by stretching faces to fake the standard height ratio, causing rejection because e-gate verifiers check landmark geometry consistency rather than visual plausibility.

**What is the specific geometric measurement required for crown-to-chin distance in a compliant passport photo?**

The vertical crown-to-menton distance must occupy exactly the standard range of the total image height as defined by ICAO Doc 9303 Part 1.

**How much does basic professional photo retouching cost on average in 2026?**

Professional basic photo retouching costs $0.50-$2 per image in the 2026 industry average according to Clipping World.

**What automated system is used to calculate facial landmark compliance quickly?**

Automated verification systems utilize the MediaPipe Face Mesh multi-point landmark model to execute this calculation very quickly.

**What is the average resubmission delay caused by head-too-small rejections in the U.K.?**

The U.K. HM Passport Office Digital Claim Service 2023 data shows an average 10-day resubmission delay for first submissions rejected with head-too-small as a top reason.

**How does incorrect face framing affect border e-gate recognition accuracy?**

NIST Face Recognition Vendor Test Part 5 2022 results demonstrate that border e-gate false non-match rises from 2.1% at compliant framing to elevated levels at below-range framing.

## Quick answers

| Why do AI apps often fail passport compliance when they stretch faces to fake the standard height ratio? | AI apps fail because e-gate verifiers check landmark geometry consistency rather than visual plausibility, causing rejection based on geometric inconsistency. |
| --- | --- |
| What is the cost range for professional basic photo retouching in 2026 according to industry averages? | Professional basic photo retouching costs $0.50-$2 per image in 2026 industry average. |
| Can beauty retouching save a passport photo if the face height is below the compliant floor? | No beauty retouch can save it because the checker rejects based on geometric inconsistency rather than aesthetic quality. |
| Which specific facial landmarks are used by automated verification systems to calculate crown-to-chin height compliance? | Automated verification systems utilize the MediaPipe Face Mesh multi-point landmark model to locate the trichion (top of the skull) and menton (lowest point of the chin). |
| What is the only valid path to resolve a non-compliant head size ratio caused by incorrect framing? | The only valid path is an optical reshot, as digital stretching or retouching cannot fix fundamental geometric mismatches. |

Also worth reading: **2026 ICAO 9303: Diffusion Headshots Fail 70% Face-Height Fix**: [2026 ICAO 9303: Diffusion Headshots](https://kahma.io/blog/2026-icao-9303-diffusion-headshots-fail-70-face-height-fix.php) · **2026 ICAO 9303: AI ID Photos Must Exceed 600x600 Pixels**: [2026 ICAO 9303: AI ID](https://kahma.io/blog/2026-icao-9303-ai-id-photos-must-exceed-600x600-pixels.php) · **600x600 Passport Spec Breaks AI Headshots: 69% Retrain Threshold**: [600x600 Passport Spec Breaks AI](https://kahma.io/blog/600x600-passport-spec-breaks-ai-headshots-69-retrain-threshold.php)

### Related reading

- [ICAO 9303: AI Passport Photos Must Hit 70–80% Head Height](https://kahma.io/blog/icao-9303-ai-passport-photos-must-hit-7080-head-height.php)
- [The 5-Minute Online Passport Photo Hack for 2024](https://kahma.io/blog/the_5_minute_online_passport_photo_hack_for_2024.php)
- [2026 ICAO 9303: Diffusion Headshots Fail 70% Face-Height Fix](https://kahma.io/blog/2026-icao-9303-diffusion-headshots-fail-70-face-height-fix.php)
- [Legal Implications of Last Month's Rent Collection 7 Key Requirements for AI-Powered Contract Analysis](https://kahma.io/blog/legal_implications_of_last_month_s_rent_collection_7_key_req.php)
- [AI-Enabled HR Platforms Impact on PHRca Recertification Credit Requirements Through 2024](https://kahma.io/blog/ai_enabled_hr_platforms_impact_on_phrca_recertification_cred.php)
- [How Part-Time Work and Education Count Towards Job Experience Requirements A Data-Driven Analysis](https://kahma.io/blog/how_part_time_work_and_education_count_towards_job_experienc.php)

### Latest

- [My Gut Friend Explained: Verify 1 Editorial Title Before Buying](https://kahma.io/blog/my-gut-friend-explained-verify-1-editorial-title-before-buying.php)
- [Passport Photos Failing Checks: 2026 Standard 35% Reshoot vs Retouch](https://kahma.io/blog/passport-photos-failing-checks-2026-standard-35-reshoot-vs-retouch.php)
- [Passport photos fail test: 34% pass Passport Standard (ICAO 9303) 2026](https://kahma.io/blog/passport-photos-fail-test-34-pass-passport-standard-icao-9303-2026.php)
- [Passport Photos Rejected 2026: 12% Head Size Fix or Retake](https://kahma.io/blog/passport-photos-rejected-2026-12-head-size-fix-or-retake.php)

Canonical: https://kahma.io/blog/passport-photo-requirements-2026-70-80-face-height-retouch-vs-reshoot.php
Markdown: https://kahma.io/blog/passport-photo-requirements-2026-70-80-face-height-retouch-vs-reshoot.php/index.md