J6E/M Benchmark of Model Performance
-
Test Board: J6E
-
Range: Models under the
samples/ucp_tutorial/dnn/ai_benchmark/j6path in the OE package -
Runtime Environment: Linux
MODEL ACCURACY
| MODEL NAME | INPUT SIZE | ACCURACY | Dataset |
|---|---|---|---|
| ResNet50 | 1x3x224x224 | Top1: 0.7704(FLOAT)/0.7665(INT8) | ImageNet |
| GoogleNet | 1x3x224x224 | Top1: 0.7018(FLOAT)/0.6998(INT8) | ImageNet |
| EfficientNet_Lite1 | 1x240x240x3 | Top1: 0.7652(FLOAT)/0.7614(INT8) | ImageNet |
| EfficientNet_Lite2 | 1x260x260x3 | Top1: 0.7734(FLOAT)/0.7697(INT8) | ImageNet |
| EfficientNet_Lite3 | 1x280x280x3 | Top1: 0.7917(FLOAT)/0.7896(INT8) | ImageNet |
| EfficientNet_Lite4 | 1x300x300x3 | Top1: 0.8063(FLOAT)/0.8043(INT8) | ImageNet |
| Vargconvnet | 1x3x224x224 | Top1: 0.7793(FLOAT)/0.7770(INT8) | ImageNet |
| Efficientnasnet_m | 1x3x300x300 | Top1: 0.7935(FLOAT)/0.7923(INT8) | ImageNet |
| Efficientnasnet_s | 1x3x280x280 | Top1: 0.7441(FLOAT)/0.7524(INT8) | ImageNet |
| ResNet18 | 1x3x224x224 | Top1: 0.6976(FLOAT)/0.6938(INT8) | ImageNet |
| YOLOv2_Darknet19 | 1x3x608x608 | [IoU=0.50:0.95]= 0.2760(FLOAT)/0.2700(INT8) | COCO |
| YOLOv3_Darknet53 | 1x3x416x416 | [IoU=0.50:0.95]= 0.3370(FLOAT)/0.3360(INT8) | COCO |
| YOLOv5x_v2.0 | 1x3x672x672 | [IoU=0.50:0.95]= 0.4810(FLOAT)/0.4670(INT8) | COCO |
| Centernet_resnet101 | 1x3x512x512 | [IoU=0.50:0.95]= 0.3420(FLOAT)/0.3270(INT8) | COCO |
| YOLOv3_VargDarknet | 1x3x416x416 | [IoU=0.50:0.95]= 0.3280(FLOAT)/0.3270(INT8) | COCO |
| Deeplabv3plus_efficientnetb0 | 1x3x1024x2048 | mIoU: 0.7630(FLOAT)/0.7571(INT8) | Cityscapes |
| Fastscnn_efficientnetb0 | 1x3x1024x2048 | mIoU: 0.6997(FLOAT)/0.6914(INT8) | Cityscapes |
| Deeplabv3plus_efficientnetm1 | 1x3x1024x2048 | mIoU: 0.7794(FLOAT)/0.7754(INT8) | Cityscapes |
| Deeplabv3plus_efficientnetm2 | 1x3x1024x2048 | mIoU: 0.7882(FLOAT)/0.7854(INT8) | Cityscapes |
| Bev_lss_efficientnetb0_multitask | image: 6x3x256x704 points(0&1): 10x128x128x2 | NDS: 0.3006(FLOAT)/0.3003(INT8) MeanIOU: 0.5180(FLOAT)/0.5171(INT8) mAP: 0.2061(FLOAT)/0.2046(INT8) | Nuscenes |
| Detr3d_efficientnetb3 | coords(0-3): 6x4x256x2 image: 6x3x512x1408 masks: 1x4x256x24 | NDS: 0.3304(FLOAT)/0.3309(INT8) mAP: 0.2753(FLOAT)/0.2743(INT8) | Nuscenes |
| Petr_efficientnetb3 | image: 6x3x512x1408 pos_embed: 1x96x44x256 | NDS: 0.3765(FLOAT)/0.3745(INT8) mAP: 0.3038(FLOAT)/0.2937(INT8) | Nuscenes |
| Bevformer_tiny_resnet50_detection | img: 6x3x480x800 prev_bev: 1x2500x256 prev_bev_ref: 1x50x50x2 queries_rebatch_grid: 6x20x32x2 restore_bev_grid: 1x100x50x2 reference_points_rebatch: 6x640x4x2 bev_pillar_counts: 1x2500x1 | NDS: 0.3713(FLOAT)/0.3695(INT8) mAP: 0.2673(FLOAT)/0.2645(INT8) | Nuscenes |
| Flashocc_henet_lss_occ3d_nuscenes | img: 6x3x512x960 points: 10x128x128x2 points_depth: 10x128x128x2 | mIoU: 0.3675(FLOAT)/0.3687(INT8) | Nuscenes |
| Centerpoint_pointpillar | points: 300000x5 voxel_feature: 1x5x20x40000 coors: 40000x4 | NDS: 0.5832(FLOAT)/0.5816(INT8) mAP: 0.4804(FLOAT)/0.4784(INT8) | Nuscenes |
| FCOS3D_efficientnetb0 | 1x3x512x896 | NDS: 0.3061(FLOAT)/0.3026(INT8) mAP: 0.2133(FLOAT)/0.2094(INT8) | nuscenes |
| Ganet_mixvargenet | 1x3x320x800 | F1Score: 0.7949(FLOAT)/0.7883(INT8) | CuLane |
| Deformable_detr_resnet50 | 1x3x800x1333 | [IoU=0.50:0.95]= 0.4413(FLOAT)/0.4497(INT8) | MS COCO |
| Unet_mobilenetv1 | 1x3x1024x2048 | mIoU: 0.6801(FLOAT)/0.6767(INT8) | Cityscapes |
| Maptroe_henet_tinym_bevformer | img: 6x3x480x800 osm_mask: 1x1x50x100 queries_rebatch_grid: 6x20x100x2 restore_bev_grid: 1x100x100x2 reference_points_rebatch: 6x2000x4x2 bev_pillar_counts: 1x5000x1 | mAP: 0.6633(FLOAT)/0.6568(INT8) | Nuscenes |
| Qcnet_oe | valid_mask: 1x30x10 valid_mask_a2a: 1x10x30x30 agent_type: 1x30x1 x_a_cur: 1x1x30x1,1x1x30x1,1x1x30x1,1x1x30x1 r_pl2a_cur: 1x1x30x80,1x1x30x80,1x1x30x80 r_t_cur: 1x1x30x6,1x1x30x6,1x1x30x6,1x1x30x6 r_a2a_cur: 1x1x30x30,1x1x30x30,1x1x30x30 x_a_mid_emb: 1x30x2x128 x_a: 1x30x6x128 pl_type,is_intersection: 1x80 r_pl2pl: 1x1x80x80,1x1x80x80,1x1x80x80 r_pt2pl: 1x1x80x50,1x1x80x50,1x1x80x50 mask_pl2pl: 1x80x80 magnitude,pt_type,side,mask: 1x80x50 mask_a2m: 1x30x30 mask_dst: 1x30x1 type_pl2pl: 1x80x80 | hitrate: 0.8026(FLOAT)/0.7982(INT8) | Argoverse 2 |
MODEL PERFORMANCE
-
FPS = Frames per second. This data is obtained by running the
hrt_model_exectool in multi-threaded mode. For detailed usage instructions, please refer to the the hrt_model_exec tool introduction - model performance evaluation section. -
Latency = Inference latency, with the unit of ms(milliseconds). This data is obtained by running the
hrt_model_exectool in single-threaded mode. For detailed usage instructions, please refer to the the hrt_model_exec tool introduction - model performance evaluation section.
| MODEL NAME | INPUT SIZE | Latency(ms) | FPS | FPS Configuration |
|---|---|---|---|---|
| ResNet50 | 1x3x224x224 | 1.148 | 1156.52 | thread_num:4 |
| GoogleNet | 1x3x224x224 | 0.626 | 2849.10 | thread_num:4 |
| EfficientNet_Lite1 | 1x240x240x3 | 0.610 | 3239.57 | thread_num:4 |
| EfficientNet_Lite2 | 1x260x260x3 | 0.707 | 2477.68 | thread_num:4 |
| EfficientNet_Lite3 | 1x280x280x3 | 0.833 | 1900.66 | thread_num:4 |
| EfficientNet_Lite4 | 1x300x300x3 | 1.070 | 1303.55 | thread_num:4 |
| Vargconvnet | 1x3x224x224 | 0.969 | 1455.76 | thread_num:4 |
| Efficientnasnet_m | 1x3x300x300 | 0.966 | 1455.76 | thread_num:4 |
| Efficientnasnet_s | 1x3x280x280 | 0.590 | 3207.72 | thread_num:4 |
| ResNet18 | 1x3x224x224 | 0.688 | 2386.00 | thread_num:4 |
| YOLOv2_Darknet19 | 1x3x608x608 | 4.671 | 227.67 | thread_num:4 |
| YOLOv3_Darknet53 | 1x3x416x416 | 5.049 | 209.86 | thread_num:4 |
| YOLOv5x_v2.0 | 1x3x672x672 | 16.290 | 62.36 | thread_num:4 |
| Centernet_resnet101 | 1x3x512x512 | 5.638 | 186.73 | thread_num:4 |
| YOLOv3_VargDarknet | 1x3x416x416 | 3.561 | 306.42 | thread_num:4 |
| Deeplabv3plus_efficientnetb0 | 1x3x1024x2048 | 6.899 | 150.81 | thread_num:4 |
| Fastscnn_efficientnetb0 | 1x3x1024x2048 | 4.368 | 244.50 | thread_num:4 |
| Deeplabv3plus_efficientnetm1 | 1x3x1024x2048 | 11.193 | 91.46 | thread_num:4 |
| Deeplabv3plus_efficientnetm2 | 1x3x1024x2048 | 15.775 | 64.40 | thread_num:4 |
| Bev_lss_efficientnetb0_multitask | image: 6x3x256x704 points(0&1): 10x128x128x2 | 5.825 | 186.28 | thread_num:4 |
| Detr3d_efficientnetb3 | coords(0-3): 6x4x256x2 image: 6x3x512x1408 masks: 1x4x256x24 | 30.994 | 32.50 | thread_num:4 |
| Petr_efficientnetb3 | image: 6x3x512x1408 pos_embed: 1x96x44x256 | 48.203 | 20.85 | thread_num:4 |
| Bevformer_tiny_resnet50_detection | img: 6x3x480x800 prev_bev: 1x2500x256 prev_bev_ref: 1x50x50x2 queries_rebatch_grid: 6x20x32x2 restore_bev_grid: 1x100x50x2 reference_points_rebatch: 6x640x4x2 bev_pillar_counts: 1x2500x1 | 31.957 | 31.53 | thread_num:4 |
| Flashocc_henet_lss_occ3d_nuscenes | img: 6x3x512x960 points: 10x128x128x2 points_depth: 10x128x128x2 | 10.970 | 93.61 | thread_num:6 |
| Centerpoint_pointpillar | points: 300000x5 voxel_feature: 1x5x20x40000 coors: 40000x4 | 12.824 | 122.84 | thread_num:4 |
| FCOS3D_efficientnetb0 | 1x3x512x896 | 2.663 | 449.64 | thread_num:4 |
| Ganet_mixvargenet | 1x3x320x800 | 0.927 | 1578.03 | thread_num:4 |
| Deformable_detr_resnet50 | 1x3x800x1333 | 200.036 | 5.01 | thread_num:6 |
| Unet_mobilenetv1 | 1x3x1024x2048 | 1.611 | 811.54 | thread_num:4 |
| Maptroe_henet_tinym_bevformer | img: 6x3x480x800 osm_mask: 1x1x50x100 queries_rebatch_grid: 6x20x100x2 restore_bev_grid: 1x100x100x2 reference_points_rebatch: 6x2000x4x2 bev_pillar_counts: 1x5000x1 | 14.481 | 70.53 | thread_num:4 |
| Qcnet_oe | valid_mask: 1x30x10 valid_mask_a2a: 1x10x30x30 agent_type: 1x30x1 x_a_cur: 1x1x30x1,1x1x30x1,1x1x30x1,1x1x30x1 r_pl2a_cur: 1x1x30x80,1x1x30x80,1x1x30x80 r_t_cur: 1x1x30x6,1x1x30x6,1x1x30x6,1x1x30x6 r_a2a_cur: 1x1x30x30,1x1x30x30,1x1x30x30 x_a_mid_emb: 1x30x2x128 x_a: 1x30x6x128 pl_type,is_intersection: 1x80 r_pl2pl: 1x1x80x80,1x1x80x80,1x1x80x80 r_pt2pl: 1x1x80x50,1x1x80x50,1x1x80x50 mask_pl2pl: 1x80x80 magnitude,pt_type,side,mask: 1x80x50 mask_a2m: 1x30x30 mask_dst: 1x30x1 type_pl2pl: 1x80x80 | 4.757 | 235.41 | thread_num:4 |
