Neuralet > Models > Jetson TX2
Jetson TX2

Jetson TX2

41

Image Classification

  • Model
  • Task
  • Inference Time (ms)
  • Throughput (FPS)
  • Description
  • File
  • Repo or Website

Object Detection

  • Model
  • Task
  • Inference Time (ms)
  • Throughput (FPS)
  • Description
  • File
  • Repo or Website
  • SSD_MobileNetV1_COCO

    72.9

    Official object detector model by NVIDIA pre-trained on the MSCOCO dataset on an input size of 300x300 that recognizes 90 classes.

    9

  • FaceNet-120

    A single-class facial recognition detector by NVIDIA.

    17

  • MultiPed-500

    A multi-class pedestrian and baggage detector by NVIDIA.

    11

  • Ped-100

    A single-class pedestrian detector by NVIDIA.

    11

  • DetectNet-COCO-Airplane

    Official object detector model by NVIDIA pre-trained on the MSCOCO dataset's airplane class that recognizes airplanes.

    14

  • DetectNet-COCO-Chair

    Official object detector model by NVIDIA pre-trained on the MSCOCO dataset's chair class that recognizes chairs.

    12

  • DetectNet-COCO-Bottle

    Official object detector model by NVIDIA pre-trained on the MSCOCO dataset's bottle class that recognizes bottles.

    9

  • DetectNet-COCO-Dog

    Official object detector model by NVIDIA pre-trained on the MSCOCO dataset's dog class that recognizes dogs.

    11

  • SSD_MobileNetV1_COCO

    Official object detector model by NVIDIA pre-trained on the MSCOCO dataset on an input size of 300x300 that recognizes 90 classes.

    10

  • SSD_InceptionV2_COCO

    132

    Official object detector model by NVIDIA pre-trained on the MSCOCO dataset on an input size of 300x300 that recognizes 90 classes.

    8

Pose Estimation

  • Model
  • Task
  • Inference Time (ms)
  • Throughput (FPS)
  • Description
  • File
  • Repo or Website
  • TRTPose_ResNet18

    Official pose estimation model by NVIDIA pre-trained on the MSCOCO dataset on an input resolution of 224x224.

    6

  • TRTPose_DenseNet121

    Official pose estimation model by NVIDIA pre-trained on the MSCOCO dataset on an input resolution of 256x256.

    7

  • OpenPifPaf_ResNet50_481x321

    Neuralet’s TensorRT-optimized version of OpenPifPaf pose estimation model pre-trained on the MSCOCO dataset on an input resolution of 481x321.

    22

  • OpenPifPaf_ResNet50_257x193

    Neuralet’s TensorRT-optimized version of OpenPifPaf pose estimation model pre-trained on the MSCOCO dataset on an input resolution of 257x193.

    22

Semantic Segmentation

  • Model
  • Task
  • Inference Time (ms)
  • Throughput (FPS)
  • Description
  • File
  • Repo or Website

Get In Touch

Have a question? Send us a message and we will respond as soon as possible.