| 模型名字 | 模型ID | 调用接口 |
|---|---|---|
| yolov3.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV3 | CVI_TDL_Detection |
| yolov5m.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV5 | CVI_TDL_Detection |
| yolov5s.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV5 | CVI_TDL_Detection |
| yolov6m.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV6 | CVI_TDL_Detection |
| yolov6s.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV6 | CVI_TDL_Detection |
| yolox_m.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOX | CVI_TDL_Detection |
| yolox_s.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOX | CVI_TDL_Detection |
| yolov7-tiny.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV7 | CVI_TDL_Detection |
| yolov8n.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV8_DETECTION | CVI_TDL_Detection |
| yolov8s.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV8_DETECTION | CVI_TDL_Detection |
| ppyoloe.cvimodel | CVI_TDL_SUPPORTED_MODEL_PPYOLOE | CVI_TDL_Detection |
| yolov10n.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV10_DETECTION | CVI_TDL_Detection |
| hardhat_detection_v2.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV8_HARDHAT | CVI_TDL_Detection |
| hand_det_qat_640x384.cvimodel | CVI_TDL_SUPPORTED_MODEL_HAND_DETECTION | CVI_TDL_Detection |
| pet_det_640x384.cvimodel | CVI_TDL_SUPPORTED_MODEL_PERSON_PETS_DETECTION | CVI_TDL_Detection |
| yolov8n_384_640_person_vehicle.cvimodel | CVI_TDL_SUPPORTED_MODEL_PERSON_VEHICLE_DETECTION | CVI_TDL_Detection |
| meeting_det_640x384.cvimodel | CVI_TDL_SUPPORTED_MODEL_HAND_FACE_PERSON_DETECTION | CVI_TDL_Detection |
| yolov8n_headperson.cvimodel | CVI_TDL_SUPPORTED_MODEL_HEAD_PERSON_DETECTION | CVI_TDL_Detection |
| mobiledetv2-d0-ls.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_COCO80 | CVI_TDL_Detection |
| mobiledetv2-d1-ls.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_COCO80 | CVI_TDL_Detection |
| mobiledetv2-d2-ls.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_COCO80 | CVI_TDL_Detection |
| mobiledetv2-vehicle-d0-ls.cvimode | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_VEHICLE | CVI_TDL_Detection |
| mobiledetv2-pedestrian-d0-ls-384.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_PEDESTRIAN | CVI_TDL_Detection |
| mobiledetv2-pedestrian-d0-ls-448.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_PEDESTRIAN | CVI_TDL_Detection |
| mobiledetv2-pedestrian-d0-ls-640.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_PEDESTRIAN | CVI_TDL_Detection |
| mobiledetv2-pedestrian-d0-ls-768.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_PEDESTRIAN | CVI_TDL_Detection |
| mobiledetv2-pedestrian-d1-ls.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_PEDESTRIAN | CVI_TDL_Detection |
| mobiledetv2-pedestrian-d1-ls-1024.cvimodel | CVI_TDL_SUPPORTED_MODEL_MOBILEDETV2_PEDESTRIAN | CVI_TDL_Detection |
| retinaface_mnet0.25_342_608.cvimodel | CVI_TDL_SUPPORTED_MODEL_RETINAFACE | CVI_TDL_FaceDetection |
| retinaface_mnet0.25_608_342.cvimodel | CVI_TDL_SUPPORTED_MODEL_RETINAFACE | CVI_TDL_FaceDetection |
| retinaface_mnet0.25_608.cvimodel | CVI_TDL_SUPPORTED_MODEL_RETINAFACE | CVI_TDL_FaceDetection |
| scrfd_320_256_ir.cvimodel | CVI_TDL_SUPPORTED_MODEL_SCRFDFACE | CVI_TDL_FaceDetection |
| scrfd_480_270_int8.cvimodel | CVI_TDL_SUPPORTED_MODEL_SCRFDFACE | CVI_TDL_FaceDetection |
| scrfd_480_360_int8.cvimodel | CVI_TDL_SUPPORTED_MODEL_SCRFDFACE | CVI_TDL_FaceDetection |
| scrfd_500m_bnkps_432_768.cvimodel | CVI_TDL_SUPPORTED_MODEL_SCRFDFACE | CVI_TDL_FaceDetection |
| retinafaceIR_mnet0.25_342_608.cvimodel | CVI_TDL_SUPPORTED_MODEL_RETINAFACE_IR | CVI_TDL_FaceDetection |
| retinafaceIR_mnet0.25_608_342.cvimodel | CVI_TDL_SUPPORTED_MODEL_RETINAFACE_IR | CVI_TDL_FaceDetection |
| retinafaceIR_mnet0.25_608_608.cvimodel | CVI_TDL_SUPPORTED_MODEL_RETINAFACE_IR | CVI_TDL_FaceDetection |
| retinaface_yolox_fdmask.cvimodel | CVI_TDL_SUPPORTED_MODEL_FACEMASKDETECTION | CVI_TDL_FaceDetection |
| pipnet_blurness_v5_64_retinaface_50ep.cvimodel | CVI_TDL_SUPPORTED_MODEL_FACELANDMARKERDET2 | CVI_TDL_FaceLandmarkerDet2 |
| fqnet-v5_shufflenetv2-softmax.cvimodel | CVI_TDL_SUPPORTED_MODEL_FACEQUALITY | CVI_TDL_FaceQuality |
| mask_classifier.cvimodel | CVI_TDL_SUPPORTED_MODEL_MASKCLASSIFICATION | CVI_TDL_MaskClassification |
| cviface-v5-m.cvimodel | CVI_TDL_SUPPORTED_MODEL_FACERECOGNITION | CVI_TDL_FaceRecognition |
| cviface-v6-s.cvimodel | CVI_TDL_SUPPORTED_MODEL_FACERECOGNITION | CVI_TDL_FaceRecognition |
| face_attr_112_112.cvimodel | CVI_TDL_SUPPORTED_MODEL_FACEATTRIBUTE_CLS | CVI_TDL_FaceAttribute_cls |
| hand_cls_128x128.cvimodel | CVI_TDL_SUPPORTED_MODEL_HANDCLASSIFICATION | CVI_TDL_HandClassification |
| hand_kpt_128x128.cvimodel | CVI_TDL_SUPPORTED_MODEL_HAND_KEYPOINT | CVI_TDL_HandKeypoint |
| hand_kpt_cls9.cvimodel | CVI_TDL_SUPPORTED_MODEL_HAND_KEYPOINT_CLASSIFICATION | CVI_TDL_HandKeypointClassification |
| wpodnet_v0_bf16.cvimodel | CVI_TDL_SUPPORTED_MODEL_WPODNET | CVI_TDL_LicensePlateDetection |
| lprnet_v0_tw_bf16.cvimodel | CVI_TDL_SUPPORTED_MODEL_LPRNET_TW | CVI_TDL_LicensePlateDetection_TW |
| lprnet_v1_cn_bf16.cvimodel | CVI_TDL_SUPPORTED_MODEL_LPRNET_CN | CVI_TDL_LicensePlateDetection_CN |
| ir_liveness.cvimodel | CVI_TDL_IrLiveness | CVI_TDL_SUPPORTED_MODEL_IRLIVENESS |
| yolov8n_pose_384_640.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV8POSE | CVI_TDL_PoseDetection |
| simcc_mv2_pose.cvimodel | CVI_TDL_SUPPORTED_MODEL_SIMCC_POSE | CVI_TDL_PoseDetection |
| yolov8n_seg.cvimodel | CVI_TDL_SUPPORTED_MODEL_YOLOV8_SEG | CVI_TDL_YoloV8_Seg |
| face_landmark_bf16.cvimodel | CVI_TDL_FaceLandmarker | CVI_TDL_SUPPORTED_MODEL_FACELANDMARKER |
| eye_v1_bf16.cvimodel | CVI_TDL_EyeClassification | CVI_TDL_SUPPORTED_MODEL_EYECLASSIFICATION |
| yawn_v1_bf16.cvimodel | CVI_TDL_YawnClassification | CVI_TDL_SUPPORTED_MODEL_YAWNCLASSIFICATION |
| c10_lightv2_mse40_mix.cvimodel | CVI_TDL_SUPPORTED_MODEL_SOUNDCLASSIFICATION | CVI_TDL_SoundClassification |
| 模型名字 | ION/FlASH | 推理耗时（ms） | 模型性能 |
|---|---|---|---|
| c10_lightv2_mse40_mix.cvimodel | 0.62 MB/587KB | 3.12 | Acc: 98.9% |
| cviface-v6-s.cvimodel | 2.97 MB/2.36MB | 7.96 | FMR:0.1 FNMR:0.0141 |
| hand_det_qat_640x384.cvimodel | 2.82MB/913KB | 16.6 | mAP(0.5): 82.1% |
| hand_kpt_128x128.cvimodel | 0.84MB/809KB | 0.829 | [email protected]:0.886 |
| hand_kpt_cls9.cvimodel | 0.05MB/54KB | 0.255 | Acc: 91.0% |
| hardhat_detection_v2.cvimodel | 7.48MB/1.05MB | 38.4 | mAP(0.5)=92.42% |
| mask_classifier.cvimodel | 3.11MB/2. 30MB | 4.88 | Acc: 97.2% |
| mobiledetv2-pedestrian-d0-ls-448.cvimodel | 2.12MB/441KB | 10.3 | mAP(0.5): 66.4% |
| pet_det_640x384.cvimodel | 6.5MB/2.99MB | 32.8 | mAP(0.5):87.0% |
| ppyoloe.cvimodel | 14.55MB/8.9MB | 101.15 | mAP(0.5): 55.4% |
| scrfd_768_432_int8_1x.cvimodel | 4.50 MB/742KB | 10.9 | mAP(0.5)：easy: 89.4% medium: 86.5% hard:65.9% |
| yolov7-tiny.cvimodel | 70.66MB/7.7MB | 70.41 | mAP(0.5): 53.4% |
| yolov8n_384_640_person_vehicle.cvimodel | 5.58 MB/3.13M | 28.5 | mAP(0.5)：72.0% |
| yolov8n.cvimodel | 31.56MB/3.5MB | 45.62 | mAP(0.5): 51.2% |
| yolov8n_headperson.cvimodel | 5.29M/3.13M | 26.5 | mAP(0.5): 78.5% |
| yolox_s.cvimodel | 95.44MB/10.0MB | 127.91 | mAP(0.5): 52.4% |
| 模型评测信息待更新 |