MotoGP weekends mean three classes, separate starting lists, and numbers hidden by lean angles. Here's the workflow that manages 100+ riders across all three categories in less time than one manual session.
MotoGP photography is deceptively harder than road racing. You're shooting three championship classes (MotoGP, Moto2, Moto3) simultaneously, with overlapping rider counts, numbers obscured by extreme lean angles, and rain conditions that spray completely obscures the fairing. The 20+ rounds from March to November mean this workflow repeats every two weeks.
Typical Event
3-day weekend (Friday: FP1/FP2 → Saturday: FP3/qualifying → Sunday: race across 3 classes)
Photo Volume
2,000-4,000 RAW files across FP, qualifying, and race sessions — roughly split 1,200 MotoGP, 1,200 Moto2, 1,200 Moto3
Delivery
Per-class delivery often within 1 hour of class finish (not waiting for all 3), requiring separate tagging batches
Key Challenge
Lean angle occlusion — at 65+ degrees, the fairing twists and the number disappears from view. Corner apex shots are hardest; entry/exit are cleaner.
Download the official three CSVs (MotoGP, Moto2, Moto3) from Dorna/FIM. Import each into RaceTagger as a separate 'class' — the AI will later identify which class each photo belongs to based on rider and bike visual features.
Pro tip
Name your CSVs clearly: motogp-2026-round-x.csv, moto2-2026-round-x, moto3-2026-round-x. When you tag later, you'll filter results by class automatically.
Shoot each session independently. After FP1, ingest with Photo Mechanic, cull heavily, then run RaceTagger on just that session's photos. The AI processes and detects class + rider number simultaneously.
Pro tip
At high lean angles, prioritize entry and exit corners for clean number visibility. Apex shots look amazing editorially but are hardest for the AI to read — plan accordingly if you're chasing perfect frames.
Run RaceTagger on your culled session folder. The AI detects both the rider number AND class simultaneously. Photos are tagged to the correct MotoGP/Moto2/Moto3 CSV entry. Multi-rider detection handles slipstream/drafting shots.
Pro tip
Process sessions immediately after the shoot. Don't wait until end-of-day to batch everything together — you'll need those MotoGP results for early delivery anyway.
RaceTagger flags low-confidence detections (rain, extreme lean, backlit shots). Additionally, review photos on the class boundary — sometimes Moto2 and Moto3 bikes can look similar, especially in wet conditions. Usually 8-12% of the total set.
Pro tip
Wet sessions will have higher flag rates. Budget 15-20% manual review in rain. In dry, you'll be closer to 5-8%.
RaceTagger writes XMP files with rider name, number, team, and CLASS field. Filter your results: export MotoGP photos as 'MotoGP-round-x', Moto2 as 'Moto2-round-x', Moto3 as 'Moto3-round-x'. One export per class.
Pro tip
Create separate Lightroom catalogs for each class if you're managing archives across the season. Keeps class galleries organized and speeds up next-round lookups.
Import each class folder into Lightroom separately. Filter by rider name, review your best shots per rider, edit for style consistency, and deliver the class gallery before the next session starts. Moto3 finishes early — you have their images live before Moto2 race even ends.
Pro tip
Per-class delivery wins contracts. Race organizers remember which photographer had MotoGP images up within 90 minutes of checkered. You'll be booked for every round.
Why it's hard: At extreme lean, the bike compresses visually and the fairing twists toward the ground. The number, painted on the side panel, rotates away from the camera. Apex shots at 65+ degrees show the bike bottom, not the painted number.
How AI helps: The AI reads numbers in context of bike shape and leaning geometry. It identifies partial numbers, understands foreshortening, and uses rider position + helmet colors to cross-reference detection.
Why it's hard: MotoGP numbers are typically 1-99 (but historic #46 in 2024), Moto2 numbers 0-99, Moto3 numbers 0-99 — overlaps exist. A #23 could be any class. Without visual context, you can't distinguish.
How AI helps: The AI trained on the distinct visual differences between bike classes — fairing shape, engine cover style, tire sizes, and overall proportions. When a #23 appears, the model identifies the class from bike geometry, not just the number.
Why it's hard: In photos where the bike is trailing another rider or mostly out of frame, only the helmet and rider gear are visible. Number isn't visible at all — you're relying on helmet color and suit combination.
How AI helps: RaceTagger's AI can identify riders by helmet/suit color patterns. While not as reliable as number detection, it provides a secondary identification for edge cases. Manual review catches these edge shots.
Why it's hard: Wet-weather MotoGP produces massive rooster tails of spray, especially on wide main straights. Water droplets on your lens + spray from other bikes = visibility drops to 20% or less.
How AI helps: The AI is honest about low-confidence in heavy spray. Rather than guessing and creating false positives, it flags the photo for manual review. You'll spend 15-20% of wet race time on manual tagging instead of 5-8% on dry.
Why it's hard: After accidents or near-misses, riders return with gravel dust caked on the fairing. The dust settles in horizontal lines and can obscure or blur the number from a distance shot.
How AI helps: The AI understands that a slightly dirty or partially obscured number is still readable if the core digits are visible. It uses context and partial information rather than requiring perfect clarity.
Manual Tagging
8-12 hours per weekend (considering 3 classes, separate sessions)
80-88% — drops significantly during rain sessions and at extreme lean angles
With RaceTagger AI
12-15 minutes processing per session + 15-20 min manual review (vs 2-3 hours manual per session)
95%+ on clean dry shots, 88-92% in rain with intelligent flagging
Real-world scenario
It's Sunday at Mugello, Italy. FP3 wrapped at 9:35 AM, you downloaded the three CSVs during breakfast. Qualifying starts at 10:15. You shot 800 photos in FP3, culled to 400 in Photo Mechanic (8 minutes), then ran RaceTagger (6 minutes). All three classes auto-tagged. While Moto3 qualifying is happening, you reviewed the 12 flagged photos from FP3 (rain was spotty), fixed three class misidentifications, and exported the three separate galleries. By 11:30 AM, your FP3 results are in Lightroom, filtered by rider. After Moto3 qualifying finishes at 11:50, you shoot, cull to 250, tag with RaceTagger (4 minutes). By 12:15 PM, Moto3 qualifying is ready for delivery. Agencies are requesting specific riders for web while MotoGP is still on track.
Your complete Moto3 qualifying gallery (25 of the best riders) is delivered at 12:30 PM. Competitors are still processing all 800 raw files. You're already shooting MotoGP qualifying. Three weeks later, you're booked as the exclusive photographer for all remaining MotoGP events.
500 free tokens included. No credit card required. Upload photos from qualifying across all three classes and see how it handles the lean angle challenges.
Start tagging for free →How does RaceTagger handle MotoGP vs Moto2 vs Moto3 classification automatically?
It learns the visual differences between classes — fairing design, engine cover, tire sidewalls, overall bike proportions. You import all three CSVs at the start, and the AI identifies which class each photo belongs to based on bike geometry, then matches to the correct starting list.
What accuracy can I expect in rain races where spray obscures everything?
In heavy rain, accuracy drops to 88-92% on visible shots, and the AI appropriately flags low-confidence detections. Plan for 15-20% manual review in wet conditions vs 5-8% in dry. It's still vastly faster than shooting blind and tagging manually.
Can I deliver per-class instead of waiting for all 3 classes to finish?
Yes, that's the recommended workflow. Tag each session separately, deliver each class as it finishes. Moto3 results at 12:30 PM, Moto2 at 2 PM, MotoGP at 4 PM. Agencies love this because they can publish results immediately without waiting for the full package.
How does extreme lean angle (65+ degrees) affect detection accuracy?
At extreme lean, the number partially compresses and the fairing rotates away from the camera. The AI still reads it using context (rider position, corner geometry, partial digits visible) but flags high-confidence uncertainty. Corner entry/exit shots are cleaner — apex apex shots are hardest.
Do I need separate RaceTagger subscriptions for each class?
No. One subscription covers all three classes. You process one batch folder with mixed-class photos, and RaceTagger separates them automatically. You export per-class based on the class field in the XMP metadata.
Related
Wet Weather Motorcycle Racing: How to Shoot and Tag MotoGP in Rain5-7 wet races per MotoGP season — rain-specific workflow and accuracy expectations
Related
Lean Angle Detection in Motorcycle Photography — How AI AdaptsDeep dive into how AI handles geometry distortion at 50+ degree lean angles
Related
Multi-Class Racing Workflow: Managing Moto2, Moto3, and Superbike PhotosBroader workflow for photographers covering multiple motorcycle racing classes simultaneously