Key Features And Enhancements
This Vitis AI Library release includes the following key features and enhancements:
- New Model Libraries
- The following new model libraries are supported.
- New Model Support
-
- Added 15 new PyTorch models
- Added one new TensorFlow2 model
- New CPU Ops Supported
- Added 12 CPU Ops.
- Custom Op Registration Supported
- Custom op registration is supported.
- xdputil Tool Update
- CPU op running is now supported by xdputil.
Changes
None.
Compatibility
The Vitis™ AI Library 2.0 is tested with the following images.
- xilinx-zcu102-dpu-v2021.2-v2.0.0.img.gz
- xilinx-zcu104-dpu-v2021.2-v2.0.0.img.gz
- xilinx-kv260-dpu-v2021.2-v2.0.0.img.gz
- xilinx-vck190-dpu-v2021.2-v2.0.0.img.gz
Model Support
The following models are supported by this version of the Vitis™ AI Library.
No. | Neural Network | ZCU102/ ZCU104/ KV260 |
VCK190 | U50LV-DPUCAHX8H | U50LV/U55C-DPUCAHX8H-DWC | VCK5000-DPUCVDX8H-DWC | VCK5000-DPUCVDX8H | Application |
---|---|---|---|---|---|---|---|---|
1 | inception_resnet_v2_tf | Y | Y | Y | Y | Y | N/A | Image Classification |
2 | inception_v1_tf | Y | Y | Y | Y | Y | Y | |
3 | inception_v3_tf | Y | Y | Y | Y | Y | N/A | |
4 | inception_v4_2016_09_09_tf | Y | Y | Y | Y | Y | N/A | |
5 | mobilenet_v1_0_25_128_tf | Y | Y | N/A | Y | Y | N/A | |
6 | mobilenet_v1_0_5_160_tf | Y | Y | N/A | Y | Y | N/A | |
7 | mobilenet_v1_1_0_224_tf | Y | Y | N/A | Y | Y | N/A | |
8 | mobilenet_v2_1_0_224_tf | Y | Y | N/A | Y | Y | N/A | |
9 | mobilenet_v2_1_4_224_tf | Y | Y | N/A | Y | Y | N/A | |
10 | resnet_v1_101_tf | Y | Y | Y | Y | Y | Y | |
11 | resnet_v1_152_tf | Y | Y | Y | Y | Y | Y | |
12 | resnet_v1_50_tf | Y | Y | Y | Y | Y | Y | |
13 | vgg_16_tf | Y | Y | Y | Y | Y | N/A | |
14 | vgg_19_tf | Y | Y | Y | Y | Y | N/A | |
15 | ssd_mobilenet_v1_coco_tf | Y | Y | N/A | Y | Y | N/A | Object Detection |
16 | ssd_mobilenet_v2_coco_tf | Y | Y | N/A | Y | Y | N/A | |
17 | ssd_resnet_50_fpn_coco_tf | Y | Y | Y | Y | Y | Y | |
18 | yolov3_voc_tf | Y | Y | Y | Y | Y | Y | |
19 | mlperf_ssd_resnet34_tf | Y | Y | Y | Y | Y | Y | |
20 | resnet50 | Y | Y | Y | Y | Y | Y | Image Classification |
21 | resnet18 | Y | Y | Y | Y | Y | Y | |
22 | inception_v1 | Y | Y | Y | Y | Y | Y | |
23 | inception_v2 | Y | Y | Y | Y | Y | N/A | |
24 | inception_v3 | Y | Y | Y | Y | Y | N/A | |
25 | inception_v4 | Y | Y | Y | Y | Y | N/A | |
26 | mobilenet_v2 | Y | Y | N/A | Y | Y | N/A | |
27 | squeezenet | Y | Y | Y | Y | Y | Y | |
28 | ssd_pedestrian_pruned_0_97 | Y | Y | Y | Y | Y | Y | ADAS Pedestrian Detection |
29 | ssd_traffic_pruned_0_9 | Y | Y | Y | Y | Y | Y | Traffic Detection |
30 | ssd_adas_pruned_0_95 | Y | Y | Y | Y | Y | Y | ADAS Vehicle Detection |
31 | ssd_mobilenet_v2 | Y | Y | N/A | Y | Y | N/A | Object Detection |
32 | refinedet_pruned_0_8 | Y | Y | Y | Y | Y | Y | |
33 | refinedet_pruned_0_92 | Y | Y | Y | Y | Y | Y | |
34 | refinedet_pruned_0_96 | Y | Y | Y | Y | Y | Y | |
35 | vpgnet_pruned_0_99 | Y | Y | Y | Y | Y | Y | ADAS Lane Detection |
36 | fpn | Y | Y | Y | Y | Y | Y | ADAS Segmentation |
37 | sp_net | Y | Y | Y | Y | Y | Y | Pose Estimation |
38 | openpose_pruned_0_3 | Y | Y | Y | Y | Y | Y | |
39 | densebox_320_320 | Y | Y | Y | Y | Y | Y | Face Detection |
40 | densebox_640_360 | Y | Y | Y | Y | Y | Y | |
41 | face_landmark | Y | Y | Y | Y | Y | Y | Face Detection and Recognition |
42 | reid | Y | Y | Y | Y | Y | Y | Object tracking |
43 | multi_task | Y | Y | Y | Y | Y | Y | ADAS |
44 | yolov3_adas_pruned_0_9 | Y | Y | Y | Y | Y | Y | Object Detection |
45 | yolov3_voc | Y | Y | Y | Y | Y | Y | |
46 | yolov3_bdd | Y | Y | Y | Y | Y | Y | |
47 | yolov2_voc | Y | Y | Y | Y | Y | Y | |
48 | yolov2_voc_pruned_0_66 | Y | Y | Y | Y | Y | Y | |
49 | yolov2_voc_pruned_0_71 | Y | Y | Y | Y | Y | Y | |
50 | yolov2_voc_pruned_0_77 | Y | Y | Y | Y | Y | Y | |
51 | facerec_resnet20 | Y | Y | Y | Y | Y | N/A | Face Recognition |
52 | facerec_resnet64 | Y | Y | Y | Y | Y | N/A | |
53 | plate_detection | Y | Y | Y | Y | Y | Y | Plate Recognition |
54 | plate_recognition | Y | Y | Y | Y | Y | Y | |
55 | FPN_Res18_Medical_segmentation | Y | Y | Y | Y | Y | Y | Medical Segmentation |
56 | refinedet_baseline | Y | Y | Y | Y | Y | Y | Object Detection |
57 | resnet50_pt | Y | Y | Y | Y | Y | Y | Image Classification |
58 | squeezenet_pt | Y | Y | Y | Y | Y | Y | |
59 | inception_v3_pt | Y | Y | Y | Y | Y | N/A | |
60 |
personreid-res50_pt |
Y | Y | Y | Y | Y | N/A | Object Tracking |
61 |
facereid-large_pt |
Y | Y | Y | Y | Y | N/A | |
62 |
facereid-small_pt |
Y | Y | Y | Y | Y | N/A | |
63 |
SemanticFPN_cityscapes_pt |
Y | Y | Y | Y | Y | Y | Segmentation |
64 |
facerec-resnet20_mixed_pt |
Y | Y | Y | Y | Y | N/A | Face Recognition |
65 | face-quality_pt | Y | Y | Y | Y | Y | Y | |
66 | MT-resnet18_mixed_pt | Y | Y | N/A | N/A | N/A | N/A | ADAS |
67 | salsanext_pt | Y | Y | Y | Y | Y | Y | Point Cloud |
68 | pointpillars_kitti_12000_0_pt pointpillars_kitti_12000_1_pt |
Y | Y | N/A | N/A | N/A | N/A | |
69 | unet_chaos-CT_pt | Y | Y | Y | Y | Y | Y | CT Segmentation |
70 | FPN-resnet18_covid19-seg_pt | Y | Y | Y | Y | Y | Y | Covid-19 Segmentation |
71 | ENet_cityscapes_pt | Y | Y | Y | Y | Y | Y | Segmentation |
72 | personreid-res18_pt | Y | Y | Y | Y | Y | N/A | Object Tracking |
73 | yolov4_leaky_spp_m | Y | Y | Y | Y | Y | N/A | Object Detection |
74 | hourglass-pe_mpii | Y | Y | N/A | N/A | N/A | N/A | Pose Estimation |
75 | retinaface | Y | Y | N/A | Y | Y | N/A | Face Detection |
76 | FPN-resnet18_Endov | Y | Y | N/A | N/A | N/A | N/A | Robot Instrument Segmentation |
77 | tiny_yolov3_vmss | Y | Y | Y | Y | Y | Y | Object Detection |
78 | face-quality | Y | Y | Y | Y | Y | Y | Face Recognition |
79 | ssdlite_mobilenet_v2_coco_tf | Y | Y | N/A | Y | Y | N/A | Object Detection |
80 | ssd_inception_v2_coco_tf | Y | Y | N/A | N/A | Y | N/A | |
81 | MLPerf_resnet50_v1.5_tf | Y | Y | Y | Y | Y | Y | Image Classification |
82 | mobilenet_edge_1_0_tf | Y | Y | N/A | N/A | Y | N/A | |
83 | mobilenet_edge_0_75_tf | Y | Y | N/A | N/A | Y | N/A | |
84 | refinedet_VOC_tf | Y | Y | Y | Y | Y | Y | Object Detection |
85 | RefineDet-Medical_EDD_tf | Y | Y | Y | Y | Y | Y | Medical Detection |
86 | resnet_v2_50_tf | Y | Y | N/A | N/A | N/A | N/A | Image Classification |
87 | resnet_v2_101_tf | Y | Y | N/A | N/A | N/A | N/A | |
88 | resnet_v2_152_tf | Y | Y | N/A | N/A | N/A | N/A | |
89 | mobilenet_v2_cityscapes_tf | Y | Y | N/A | N/A | N/A | N/A | Segmentation |
90 | inception_v2_tf | Y | Y | N/A | N/A | Y | N/A | Image Classification |
91 | resnet50_tf2 | Y | Y | Y | Y | Y | Y | |
92 | mobilenet_1_0_224_tf2 | Y | Y | N/A | Y | Y | N/A | |
93 | inception_v3_tf2 | Y | Y | Y | Y | Y | N/A | |
94 | medical_seg_cell_tf2 | Y | Y | Y | Y | Y | Y | Medical Segmentation |
95 | semantic_seg_citys_tf2 | Y | Y | Y | Y | Y | Y | Segmentation |
96 | efficientNet-edgetpu-S_tf | Y | Y | N/A | N/A | Y | N/A | Image Classification |
97 | efficientNet-edgetpu-M_tf | Y | Y | N/A | N/A | Y | N/A | |
98 | efficientNet-edgetpu-L_tf | Y | Y | N/A | N/A | Y | N/A | |
99 | SemanticFPN_Mobilenetv2_pt | Y | Y | N/A | Y | Y | N/A | Segmentation |
100 | pointpillars_nuscenes_40000_64_0_pt pointpillars_nuscenes_40000_64_1_pt |
Y | Y | Y | Y | Y | N/A | 3D object detection |
101 | pointpainting_nuscenes_40000_64_0_pt pointpainting_nuscenes_40000_64_1_pt |
Y | Y | Y | Y | Y | N/A | 2D semantic segmentation and 3D object detection |
102 | salsanext_v2_pt | Y | Y | Y | Y | Y | Y | 3D Segmentation |
103 | centerpoint_0_pt centerpoint_1_pt |
Y | Y | N/A | N/A | N/A | N/A | 4D radar detection |
104 | multi_task_v3_pt | Y | Y | N/A | N/A | N/A | N/A | ADAS |
105 | FADNet_0_pt FADNet_1_pt FADNet_2_pt |
Y | Y | Y | Y | Y | Y | Depth Estimation |
106 | rcan_pruned_tf | Y | Y | N/A | N/A | N/A | N/A | Super Resolution |
107 | efficientnet_tf | N/A | Y | N/A | N/A | N/A | N/A | Classification |
108 | yolov4_leaky_spp_m_pruned_0_36 | Y | Y | Y | Y | N/A | N/A | Object Detection |
109 | pmg_pt | Y | Y | Y | Y | Y | N/A | Brand Recognition |
110 | bbc_pt | Y | Y | N/A | N/A | N/A | N/A | Bayesian Crowd Counting |
111 | SA_gate_pt | N/A | Y | N/A | N/A | N/A | N/A | Indoor Segmentation |
112 | ultrafast_pt | Y | Y | Y | Y | Y | Y | Road Line Detection |
113 | HardNet_MSeg_pt | Y | Y | N/A | N/A | N/A | N/A | Polyp Segmentation |
114 | drunet_pt | Y | Y | Y | Y | Y | Y | Image Denoising |
115 | person-orientation_pruned_558m_pt | Y | Y | Y | Y | Y | N/A | Person Orientation Estimation |
116 | ofa_resnet50_0_9B_pt | Y | Y | Y | Y | Y | N/A | Classification |
117 | SESR_S_pt | Y | Y | Y | Y | Y | Y | Image Super-Resolution |
118 | C2D2_Lite_0_pt C2D2_Lite_1_pt |
Y | Y | N/A | N/A | N/A | N/A | Coverage Prediction |
119 | ofa_depthwise_res50_pt | Y | Y | N/A | N/A | Y | N/A | Classification |
120 | FADNet_pruned_0_pt FADNet_pruned_1_pt FADNet_pruned_2_pt |
Y | Y | N/A | N/A | N/A | N/A | Depth Estimation |
121 | PSMNet_pruned_0_pt PSMNet_pruned_1_pt PSMNet_pruned_2_pt |
N/A | Y | N/A | N/A | N/A | N/A | |
122 | FairMot_pt | Y | Y | Y | Y | Y | Y | Joint detection and Tracking |
123 | mobilenet_v3_small_1_0_tf2 | Y | Y | N/A | N/A | N/A | N/A | Classification |
124 | ssr_pt | Y | Y | N/A | N/A | N/A | N/A | Spectral Remove |
125 | clocs_pointpillars_kitti_0_pt clocs_pointpillars_kitti_1_pt clocs_fusion_cnn_pt clocs_yolox_pt |
Y | Y | N/A | N/A | N/A | N/A | Image-lidar fusion 3D Detection |
126 | tsd_yolox_pt | Y | Y | Y | Y | Y | N/A | Traffic Sign Detection |
127 | solo_pt | N/A | Y | N/A | N/A | N/A | N/A | Instance Segmentation |
|
Device Support
The following platforms and evaluation boards (EVB) are supported by the Vitis™ AI Library 2.0.
Platform | EVB | Version |
---|---|---|
Zynq UltraScale+ MPSoC ZU9EG | Xilinx ZCU102 | 1.1 |
Zynq® UltraScale+™ MPSoC ZU7EV | Xilinx ZCU104 | 1.0 |
Zynq UltraScale+ MPSoC | Xilinx Kria™ KV260 | 1.0 |
Versal AI Core series VC1902 | Xilinx VCK190 | Production |
Accelerator Cards |
---|
Xilinx Alveo U50LV Data Center accelerator card |
Xilinx Alveo U200 Data Center accelerator card |
Xilinx Alveo U250 Data Center accelerator card |
Xilinx Alveo U55C Data Center accelerator card |
Versal AI Core series VCK5000 Data Center development kit |
Limitations
- For VCK5000, only production cards are supported. Use Vitis AI 1.4.x if using VCK5000-ES1 cards.
- Alveo U50 and U280 cards are no longer supported. Use Vitis AI 1.4.x if using these cards.
- Due to limitations of the Docker environment, MultiTask demos cannot run on the DRM mode on Cloud boards.