Abuudiii/fpv-perception-pipeline
object-detection model by Abuudiii
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
A quadrotor races through gates using only its FPV camera and an IMU, outputting the four stick commands a human pilot sends (thrust and roll / pitch / yaw rate). Everything is simulation. This is the follow-up to Abuudiii/fpv-gate-racing, where a single CNN flew from pixels and needed the next gate painted green on unseen tracks. Here nothing is painted: a detector finds the gates, written geometry turns them into a state estimate, and a PPO policy flies on that estimate. Code: https://github.com/Abuudiii/rl-experiments/tree/main/fpv-racing Unreal Engine 5, flown by the pipeline, no highlighted gates. Each is the first flight from a fixed start, not picked. Slalom (26 gates in 30 s, no crash) Split-S (41 gates in 30 s, no crash; the drone flicks upside down at the high gate and drops to the low one) Six tracks, 3D view of the flight paths (pipeline flights in Unreal) media/slalomdetectorview.mp4 and media/splitsdetectorview.mp4…
Read the full model card ↗ · Preview checked 2026-11-22T07:03:42.182Z
Inside the original model card — Document outline
- FPV gate racing from the camera: gate detector + geometry + PPO policy
- Videos
- Results
- How it works
- Files
- More videos and figures
- Use
- Limits
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. Reported project/model license: mit. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
Abuudiii/fpv-perception-pipeline is a object-detection model repository published by Abuudiii on Hugging Face. A library was not reported.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- object-detection
- Library
- Not reported
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- mit
- Access
- Not gated by Hugging Face
- Created
- 2026-11-22T06:48:00.000Z
- Last modified
- 2026-11-22T06:51:50.000Z
Compatibility and lineage
Language coverage was not reported.
Base-model lineage was not included.
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-11-22T07:03:33.420Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.
Related models
PekingU/rtdetr_r101vd_coco_o365
microsoft/table-transformer-structure-recognition
PaddlePaddle/PP-DocLayoutV3_safetensors