Qualcomm® AI HubAI Hub
HomeCompute ModelsFFNet-78S-Quantized

FFNet-78S-Quantized

Semantic segmentation for automotive street scenes.

FFNet-78S-Quantized is a "fuss-free network" that segments street scene images with per-pixel classes like road, sidewalk, and pedestrian. Trained on the Cityscapes dataset.

Snapdragon® X Elite
TorchScripttoONNX Runtime
9.43ms
Inference Time
6MB
Memory Usage
149NPU
Layers

Technical Details

Model checkpoint:ffnet78S_dBBB_cityscapes_state_dict_quarts
Input resolution:2048x1024
Number of parameters:27.5M
Model size:26.7 MB

Applicable Scenarios

  • Automotive
  • Autonomous Driving
  • Camera

Licenses

Source Model:BSD-3-CLAUSE
Deployable Model:AI Model Hub License

Tags

  • quantized
    A “quantized” model can run in low or mixed precision, which can substantially reduce inference latency.
  • real-time
    A “real-time” model can typically achieve 5-60 predictions per second. This translates to latency ranging up to 200 ms per prediction.

Supported Compute Chipsets

  • Snapdragon® X Elite