Predicting roulette
from video.
Roulette is deterministic, but small differences at the start of a spin compound quickly.
The project recovers that motion from video, then tests how far ahead it can be forecast.
TRACKING EXAMPLES
Select an image to see what the detector recorded.Recovering
the motion.
Each camera view has to become a clean, comparable trajectory before it can be used for prediction.
Source footage
The dataset contains 42 public videos and 5,463,775 frames, recorded across different wheels, lighting conditions, and camera angles.- 42
- videos
- 5,463,775
- frames
- 1.36 px
- centroid error
PIPELINE / 01INPUT
TRACKSMOOTHDERIVE
PULL DISTANCE0.0px
VECTOR SPEED260px/s
STATEREADY · READY
2,765 VALID SPINS
- TRAIN
- 75%
- VALIDATE
- 12.5%
- TEST
- 12.5%
INPUT
T × 11
featuresBALL TRAJECTORY
LSTM128 / 128 units
OUTPUTsin θ̂, cos θ̂
training lossvalidation loss
- OPTIMISER
- Adam
- BATCH
- 64
- TRAJECTORY LR
- 5e−4
- WEIGHT DECAY
- 1e−5
- EARLY STOPPING
- 10 epochs
SEARCH SPACE AT T = 300
25configurations
32–256hidden units
1–3layers
0.1–0.5dropout
More context,
lower error.
Across all three tasks, predictions improved as the observed window grew.
BASELINE COMPARISON AT T = 300
At six seconds of input, the ball-path model is compared with two simple baselines.
WHERE IT FAILS
Sudden slowing and missing frames still cause large errors.Read the full paper.
The dissertation contains the complete method, experiments, limitations, and results.
Read the dissertation