Question 1

What does this tool do?

Accepted Answer

Our AI analyzes trading card images to estimate: Centering, Surface condition, Edge wear, Corner wear, Overall raw card condition The result is an unofficial grading estimate designed to help collectors pre-screen cards before submitting to professional grading companies.

Question 2

Is this an official PSA/BGS/CGC grade?

Accepted Answer

No. This tool provides an AI-powered estimate only. We are not affiliated with: PSA, Beckett Grading Services, CGC Actual grades may vary depending on: grader interpretation, lighting, surface defects, print quality, handling during submission

Question 3

Which cards are supported?

Accepted Answer

Most modern trading cards are supported, including: Pokémon, Yu-Gi-Oh!, Magic: The Gathering, One Piece, Disney Lorcana, Flesh and Blood, MetaZoo, Star Wars, Final Fantasy TCG, Sorcery: Contested Realm, Gundam, Riftbound, Baseball, Basketball, Hockey, MMA Support quality may vary depending on card style, scan quality, and image resolution.

Question 4

How accurate is the grading estimate?

Accepted Answer

The AI performs best on: high-resolution scans, evenly lit images, straight cards, clean backgrounds Estimated grades should be treated as a screening tool, not a guaranteed final grade.

Question 5

What images work best?

Accepted Answer

Best results come from: flatbed scans, straight-on photos, high resolution images, neutral backgrounds, minimal glare, full card visible in frame

Question 6

Should I upload both front and back?

Accepted Answer

Yes. For the most accurate estimate, upload: front image, back image Some defects only appear on one side.

Question 7

Can I upload photos in sleeves or top loaders?

Accepted Answer

It's not recommended. Sleeves and cases can introduce: reflections, glare, warped edges, false surface defects Raw cards produce the most accurate results.

Question 8

Why does the detection box look slightly off?

Accepted Answer

The visual overlay is an approximate detection preview. In some cases: image scaling, perspective correction, internal preprocessing, browser rendering may cause the displayed box to appear slightly misaligned. This usually does not affect the actual grading analysis.

Question 9

What file types are supported?

Accepted Answer

Supported formats typically include: JPG / JPEG, PNG, WEBP For best accuracy, PNG or high-quality JPG is recommended.

Question 10

How is centering calculated?

Accepted Answer

The AI measures the spacing between: left/right borders, top/bottom borders and estimates the front and back centering ratios.

Question 11

Why is my centering estimate different from PSA?

Accepted Answer

Different grading companies use slightly different centering standards and tolerances. AI estimates are intended as guidance only.

Question 12

Does whitening affect centering?

Accepted Answer

No. Whitening affects edge/corner grades, not centering directly.

Question 13

Can the AI detect scratches and dents?

Accepted Answer

Minor surface defects may be detected, but some flaws are difficult to identify from images alone, especially: micro scratches, pressure dents, print lines, holo scratching

Question 14

Why did my card receive a lower score than expected?

Accepted Answer

Common causes: edge whitening, poor centering, corner wear, glare mistaken for defects, low image quality, compression artifacts

Question 15

Can this predict a PSA 10?

Accepted Answer

No tool can guarantee a PSA 10. Professional grading includes: in-person inspection, microscopic analysis, subjective grading decisions

Question 16

Are my uploaded images stored?

Accepted Answer

Our current policy: temporary processing only, deleted after analysis, private by default

Question 17

Do you sell uploaded images or data?

Accepted Answer

No user-uploaded images are sold to third parties.

Question 18

Why did the AI fail to detect my card?

Accepted Answer

Detection may fail due to: blurry images, cropped edges, heavy glare, dark backgrounds, extreme angles, partially visible cards

Question 19

Why are results inconsistent between uploads?

Accepted Answer

Small changes in: lighting, rotation, compression, image scaling can affect computer vision models. Using high-quality scans usually produces the most stable results.

Frequently Asked Questions

General Questions

Image Upload Questions

Centering Questions

Grading Questions

Privacy & Usage