# 對影片幀應用語義分割

在運行模型的管道命令之前，請遵循必要的 [先決條件](https://docs.qualcomm.com/doc/80-70018-15BT/topic/ai-hub-qualcomm-im-sdk.html#prerequisites)。

運行以下命令以確保結果顯示在連接的顯示器上：

export XDG_RUNTIME_DIR=/dev/socket/weston && export WAYLAND_DISPLAY=wayland-1
    Copy to clipboard

## DeepLabV3-Plus\_MobileNet-Quantized

DeepLabV3 量化設計用於多尺度語義分割，並在各種數據集上進行訓練。

AI Hub 模型基於 [此 DeepLabV3-Plus-MobileNet-Quantized 的實現](https://github.com/jfzhang95/pytorch-deeplab-xception)。

- 模型：[deeplabv3_plus_mobilenet_quantized.tflite](https://aihub.qualcomm.com/iot/models/deeplabv3_plus_mobilenet_quantized)
- 標籤：[voc_labels.txt](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/labels/voc_labels.txt)

gst-launch-1.0 -e --gst-debug=2 \
    filesrc location=/etc/media/video.mp4 ! qtdemux ! queue ! h264parse ! v4l2h264dec capture-io-mode=4 output-io-mode=4 ! video/x-raw,format=NV12 ! queue ! tee name=split \
    split. ! queue ! qtivcomposer name=mixer sink_1::alpha=0.5 ! queue ! waylandsink sync=true fullscreen=true \
    split. ! queue ! qtimlvconverter ! queue ! qtimltflite delegate=external external-delegate-path=libQnnTFLiteDelegate.so \
    external-delegate-options="QNNExternalDelegate,backend_type=htp;" model=/etc/models/deeplabv3_plus_mobilenet_quantized.tflite ! queue ! \
    qtimlvsegmentation module=deeplab-argmax labels=/etc/labels/voc_labels.txt \
    constants="deeplab,q-offsets=<92.0>,q-scales=<0.04518842324614525>;" ! video/x-raw,format=BGRA,width=256,height=144 ! queue ! mixer.
    Copy to clipboard

## FCN-Resnet50-Quantized

FCN\_ResNet50 是一個量化的機器學習模型，可以從 COCO 數據集中分割圖像。

AI Hub 模型基於 [此 FCN-ResNet50-Quantized 的實現](https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py)。

- 模型：[FCN-ResNet50-Quantized.tflite](https://aihub.qualcomm.com/iot/models/fcn_resnet50_quantized)
- 標籤：[voc_labels.txt](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/labels/voc_labels.txt)

備註

此管道目前不支持 QCS6490。

gst-launch-1.0 -e --gst-debug=2 \
    filesrc location=/etc/media/video.mp4 ! qtdemux ! queue ! h264parse ! v4l2h264dec capture-io-mode=4 output-io-mode=4 ! video/x-raw,format=NV12 ! queue ! tee name=split \
    split. ! queue ! qtivcomposer name=mixer sink_1::alpha=0.5 ! queue ! waylandsink sync=true fullscreen=true \
    split. ! queue ! qtimlvconverter ! queue ! qtimltflite delegate=external external-delegate-path=libQnnTFLiteDelegate.so \
    external-delegate-options="QNNExternalDelegate,backend_type=htp;" model=/etc/models/fcn_resnet50_quantized.tflite ! queue ! \
    qtimlvsegmentation module=deeplab-argmax labels=/etc/labels/voc_labels.txt \
    constants="deeplab,q-offsets=<0.0>,q-scales=<1.0>;" ! video/x-raw,format=BGRA,width=256,height=144 ! queue ! mixer.
    Copy to clipboard

## FFNet-40S-Quantized

FFNet-40S-Quantized 是一個“無憂網絡”，可以使用每像素類別如道路、人行道和行人分割街景圖像。

它在 cityscapes 數據集上進行訓練。

AI Hub 模型基於 [此 FFNet-40S-Quantized 的實現](https://github.com/Qualcomm-AI-research/FFNet)

- 模型：[ffnet_40s_quantized.tflite](https://aihub.qualcomm.com/iot/models/ffnet_40s_quantized)
- 標籤：[voc_labels.txt](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/labels/voc_labels.txt)

gst-launch-1.0 -e --gst-debug=2 \
    filesrc location=/etc/media/video.mp4 ! qtdemux ! queue ! h264parse ! v4l2h264dec capture-io-mode=4 output-io-mode=4 ! video/x-raw,format=NV12 ! queue ! tee name=split \
    split. ! queue ! qtivcomposer name=mixer sink_1::alpha=0.5 ! queue ! waylandsink sync=false fullscreen=true \
    split. ! queue ! qtimlvconverter ! queue ! qtimltflite delegate=external external-delegate-path=libQnnTFLiteDelegate.so \
    external-delegate-options="QNNExternalDelegate,backend_type=htp;" model=/etc/models/ffnet_40s_quantized.tflite ! queue ! \
    qtimlvsegmentation module=deeplab-argmax labels=/etc/labels/voc_labels.txt constants="ffnet,q-offsets=<178.0>,q-scales=<0.31378185749053955>;" ! \
    video/x-raw,format=BGRA,width=256,height=144 ! queue ! mixer.
    Copy to clipboard

## FFNet-54S-Quantized

FFNet-54S-Quantized 是一個“無憂網絡”，可以使用每像素類別如道路、人行道和行人 分割街景圖像。

它在 cityscapes 數據集上進行訓練。

AI Hub 模型基於 [此 FFNet-54S-Quantized 的實現](https://github.com/Qualcomm-AI-research/FFNet)。

- 模型：[ffnet_54s_quantized.tflite](https://aihub.qualcomm.com/iot/models/ffnet_54s_quantized)
- 標籤：[voc_labels.txt](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/labels/voc_labels.txt)

gst-launch-1.0 -e --gst-debug=2 \
    filesrc location=/etc/media/video.mp4 ! qtdemux ! queue ! h264parse ! v4l2h264dec capture-io-mode=4 output-io-mode=4 ! video/x-raw,format=NV12 ! queue ! tee name=split \
    split. ! queue ! qtivcomposer name=mixer sink_1::alpha=0.5 ! queue ! waylandsink sync=false fullscreen=true \
    split. ! queue ! qtimlvconverter ! queue ! qtimltflite delegate=external external-delegate-path=libQnnTFLiteDelegate.so \
    external-delegate-options="QNNExternalDelegate,backend_type=htp;" model=/etc/models/ffnet_54s_quantized.tflite ! queue ! \
    qtimlvsegmentation module=deeplab-argmax labels=/etc/labels/voc_labels.txt constants="ffnet,q-offsets=<178.0>,q-scales=<0.2929433584213257>;" ! \
    video/x-raw,format=BGRA,width=256,height=144 ! queue ! mixer.
    Copy to clipboard

## FFNet-78S-Quantized

FFNet-78S-Quantized 是一個“無憂網絡”，可以使用每像素類別如道路、人行道和行人 分割街景圖像。

它在 cityscapes 數據集上進行訓練。

AI Hub 模型基於 [此 FFNet-78S-Quantized 的實現](https://github.com/Qualcomm-AI-research/FFNet)。

- 模型：[ffnet_78s_quantized.tflite](https://aihub.qualcomm.com/iot/models/ffnet_78s_quantized)
- 標籤：[voc_labels.txt](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/labels/voc_labels.txt)

gst-launch-1.0 -e --gst-debug=2 \
    filesrc location=/etc/media/video.mp4 ! qtdemux ! queue ! h264parse ! v4l2h264dec capture-io-mode=4 output-io-mode=4 ! video/x-raw,format=NV12 ! queue ! tee name=split \
    split. ! queue ! qtivcomposer name=mixer sink_1::alpha=0.5 ! queue ! waylandsink sync=false fullscreen=true \
    split. ! queue ! qtimlvconverter ! queue ! qtimltflite delegate=external external-delegate-path=libQnnTFLiteDelegate.so \
    external-delegate-options="QNNExternalDelegate,backend_type=htp;" model=/etc/models/ffnet_78s_quantized.tflite ! queue ! \
    qtimlvsegmentation module=deeplab-argmax labels=/etc/labels/voc_labels.txt constants="ffnet,q-offsets=<171.0>,q-scales=<0.3849360942840576>;" ! \
    video/x-raw,format=BGRA,width=256,height=144 ! queue ! mixer.
    Copy to clipboard

Last Published: Oct 15, 2025

[Previous Topic
偵測對象](https://docs.qualcomm.com/bundle/publicresource/80-70018-15BT/topics/object-detection.md) [Next Topic
使用超分辨率放大圖像](https://docs.qualcomm.com/bundle/publicresource/80-70018-15BT/topics/super-resolution.md)