domains / families / data / tao-train-sparse4d

tao-train-sparse4d

separate

Prepare calibrated multi-camera data and train, evaluate, run inference, export, or quantize TAO Sparse4D 3D detection and tracking models. Includes 2D-to-3D geometric distillation with released LTT weights. Use for Sparse4D adaptation and temporal multi-camera perception workflows.

When to use

• train Sparse4D models
• evaluate Sparse4D models
• run inference
• export models
• quantize models
• 2D-to-3D geometric distillation
• temporal multi-camera perception workflows
• Sparse4D adaptation

When not to use

• reporting unpublished experimental fixes as validation
• changing architecture to make a checkpoint load partially
• using random initialization without explicit request

Skill page

Source