thebears/deepstream_yolo
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add YOLO-Seg

Browse Source
This commit is contained in:
Marcos Luciano
2023-09-07 00:25:56 -03:00
parent 0f4377f540
commit 1a9df997a4
10 changed files with 28 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@@ -4,6 +4,7 @@ NVIDIA DeepStream SDK 6.3 / 6.2 / 6.1.1 / 6.1 / 6.0.1 / 6.0 / 5.1 configuration
-------------------------------------------------------------------------------------------------- --------------------------------------------------------------------------------------------------
### YOLO-Pose: https://github.com/marcoslucianops/DeepStream-Yolo-Pose ### YOLO-Pose: https://github.com/marcoslucianops/DeepStream-Yolo-Pose
### YOLO-Seg: https://github.com/marcoslucianops/DeepStream-Yolo-Seg
-------------------------------------------------------------------------------------------------- --------------------------------------------------------------------------------------------------
### Important: please export the ONNX model with the new export file, generate the TensorRT engine again with the updated files, and use the new config_infer_primary file according to your model ### Important: please export the ONNX model with the new export file, generate the TensorRT engine again with the updated files, and use the new config_infer_primary file according to your model
-------------------------------------------------------------------------------------------------- --------------------------------------------------------------------------------------------------
@@ -29,13 +30,14 @@ NVIDIA DeepStream SDK 6.3 / 6.2 / 6.1.1 / 6.1 / 6.0.1 / 6.0 / 5.1 configuration
* INT8 calibration (PTQ) for Darknet and ONNX exported models * INT8 calibration (PTQ) for Darknet and ONNX exported models
* New output structure (fix wrong output on DeepStream < 6.2) - it need to export the ONNX model with the new export file, generate the TensorRT engine again with the updated files, and use the new config_infer_primary file according to your model * New output structure (fix wrong output on DeepStream < 6.2) - it need to export the ONNX model with the new export file, generate the TensorRT engine again with the updated files, and use the new config_infer_primary file according to your model
* **YOLO-Pose: https://github.com/marcoslucianops/DeepStream-Yolo-Pose** * **YOLO-Pose: https://github.com/marcoslucianops/DeepStream-Yolo-Pose**
* **YOLO-Seg: https://github.com/marcoslucianops/DeepStream-Yolo-Seg**
## ##
### Getting started ### Getting started
* [Requirements](#requirements) * [Requirements](#requirements)
* [Suported models](#supported-models) * [Supported models](#supported-models)
* [Benchmarks](docs/benchmarks.md) * [Benchmarks](docs/benchmarks.md)
* [dGPU installation](docs/dGPUInstalation.md) * [dGPU installation](docs/dGPUInstalation.md)
* [Basic usage](#basic-usage) * [Basic usage](#basic-usage)
@@ -156,7 +158,7 @@ NVIDIA DeepStream SDK 6.3 / 6.2 / 6.1.1 / 6.1 / 6.0.1 / 6.0 / 5.1 configuration
## ##
### Suported models ### Supported models
* [Darknet](https://github.com/AlexeyAB/darknet) * [Darknet](https://github.com/AlexeyAB/darknet)
* [MobileNet-YOLO](https://github.com/dog-qiuqiu/MobileNet-Yolo) * [MobileNet-YOLO](https://github.com/dog-qiuqiu/MobileNet-Yolo)

2
docs/DAMOYOLO.md
View File

@@ -166,6 +166,7 @@ parse-bbox-func-name=NvDsInferParseYoloE
**NOTE**: The **DAMO-YOLO** do not resize the input with padding. To get better accuracy, use **NOTE**: The **DAMO-YOLO** do not resize the input with padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=0 maintain-aspect-ratio=0
... ...
@@ -174,6 +175,7 @@ maintain-aspect-ratio=0
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...

4
docs/PPYOLOE.md
View File

@@ -145,6 +145,7 @@ parse-bbox-func-name=NvDsInferParseYoloE
**NOTE**: The **PP-YOLOE+ and PP-YOLOE legacy** do not resize the input with padding. To get better accuracy, use **NOTE**: The **PP-YOLOE+ and PP-YOLOE legacy** do not resize the input with padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=0 maintain-aspect-ratio=0
... ...
@@ -153,6 +154,7 @@ maintain-aspect-ratio=0
**NOTE**: The **PP-YOLOE+** uses zero mean normalization on the image preprocess. It is important to change the `net-scale-factor` according to the trained values. **NOTE**: The **PP-YOLOE+** uses zero mean normalization on the image preprocess. It is important to change the `net-scale-factor` according to the trained values.
``` ```
[property]
... ...
net-scale-factor=0.0039215697906911373 net-scale-factor=0.0039215697906911373
... ...
@@ -163,6 +165,7 @@ net-scale-factor=0.0039215697906911373
Default: `mean = 0.485, 0.456, 0.406` and `std = 0.229, 0.224, 0.225` Default: `mean = 0.485, 0.456, 0.406` and `std = 0.229, 0.224, 0.225`
``` ```
[property]
... ...
net-scale-factor=0.0173520735727919486 net-scale-factor=0.0173520735727919486
offsets=123.675;116.28;103.53 offsets=123.675;116.28;103.53
@@ -172,6 +175,7 @@ offsets=123.675;116.28;103.53
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...

4
docs/YOLONAS.md
View File

@@ -201,6 +201,7 @@ parse-bbox-func-name=NvDsInferParseYoloE
**NOTE**: The **YOLO-NAS** resizes the input with left/top padding. To get better accuracy, use **NOTE**: The **YOLO-NAS** resizes the input with left/top padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=1 maintain-aspect-ratio=1
symmetric-padding=0 symmetric-padding=0
@@ -210,6 +211,7 @@ symmetric-padding=0
**NOTE**: The **pre-trained YOLO-NAS** uses zero mean normalization on the image preprocess. It is important to change the `net-scale-factor` according to the trained values. **NOTE**: The **pre-trained YOLO-NAS** uses zero mean normalization on the image preprocess. It is important to change the `net-scale-factor` according to the trained values.
``` ```
[property]
... ...
net-scale-factor=0.0039215697906911373 net-scale-factor=0.0039215697906911373
... ...
@@ -218,6 +220,7 @@ net-scale-factor=0.0039215697906911373
**NOTE**: The **custom YOLO-NAS** uses no normalization on the image preprocess. It is important to change the `net-scale-factor` according to the trained values. **NOTE**: The **custom YOLO-NAS** uses no normalization on the image preprocess. It is important to change the `net-scale-factor` according to the trained values.
``` ```
[property]
... ...
net-scale-factor=1 net-scale-factor=1
... ...
@@ -226,6 +229,7 @@ net-scale-factor=1
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...

2
docs/YOLOR.md
View File

@@ -184,6 +184,7 @@ parse-bbox-func-name=NvDsInferParseYolo
**NOTE**: The **YOLOR** resizes the input with center padding. To get better accuracy, use **NOTE**: The **YOLOR** resizes the input with center padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=1 maintain-aspect-ratio=1
symmetric-padding=1 symmetric-padding=1
@@ -193,6 +194,7 @@ symmetric-padding=1
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...

4
docs/YOLOX.md
View File

@@ -150,6 +150,7 @@ parse-bbox-func-name=NvDsInferParseYolo
**NOTE**: The **YOLOX and YOLOX legacy** resize the input with left/top padding. To get better accuracy, use **NOTE**: The **YOLOX and YOLOX legacy** resize the input with left/top padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=1 maintain-aspect-ratio=1
symmetric-padding=0 symmetric-padding=0
@@ -159,6 +160,7 @@ symmetric-padding=0
**NOTE**: The **YOLOX** uses no normalization on the image preprocess. It is important to change the `net-scale-factor` according to the trained values. **NOTE**: The **YOLOX** uses no normalization on the image preprocess. It is important to change the `net-scale-factor` according to the trained values.
``` ```
[property]
... ...
net-scale-factor=1 net-scale-factor=1
... ...
@@ -169,6 +171,7 @@ net-scale-factor=1
Default: `mean = 0.485, 0.456, 0.406` and `std = 0.229, 0.224, 0.225` Default: `mean = 0.485, 0.456, 0.406` and `std = 0.229, 0.224, 0.225`
``` ```
[property]
... ...
net-scale-factor=0.0173520735727919486 net-scale-factor=0.0173520735727919486
offsets=123.675;116.28;103.53 offsets=123.675;116.28;103.53
@@ -178,6 +181,7 @@ offsets=123.675;116.28;103.53
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...

2
docs/YOLOv5.md
View File

@@ -176,6 +176,7 @@ parse-bbox-func-name=NvDsInferParseYolo
**NOTE**: The **YOLOv5** resizes the input with center padding. To get better accuracy, use **NOTE**: The **YOLOv5** resizes the input with center padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=1 maintain-aspect-ratio=1
symmetric-padding=1 symmetric-padding=1
@@ -185,6 +186,7 @@ symmetric-padding=1
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...

2
docs/YOLOv6.md
View File

@@ -176,6 +176,7 @@ parse-bbox-func-name=NvDsInferParseYolo
**NOTE**: The **YOLOv6** resizes the input with center padding. To get better accuracy, use **NOTE**: The **YOLOv6** resizes the input with center padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=1 maintain-aspect-ratio=1
symmetric-padding=1 symmetric-padding=1
@@ -185,6 +186,7 @@ symmetric-padding=1
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...

2
docs/YOLOv7.md
View File

@@ -178,6 +178,7 @@ parse-bbox-func-name=NvDsInferParseYolo
**NOTE**: The **YOLOv7** resizes the input with center padding. To get better accuracy, use **NOTE**: The **YOLOv7** resizes the input with center padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=1 maintain-aspect-ratio=1
symmetric-padding=1 symmetric-padding=1
@@ -187,6 +188,7 @@ symmetric-padding=1
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...

2
docs/YOLOv8.md
View File

@@ -169,6 +169,7 @@ parse-bbox-func-name=NvDsInferParseYolo
**NOTE**: The **YOLOv8** resizes the input with center padding. To get better accuracy, use **NOTE**: The **YOLOv8** resizes the input with center padding. To get better accuracy, use
``` ```
[property]
... ...
maintain-aspect-ratio=1 maintain-aspect-ratio=1
symmetric-padding=1 symmetric-padding=1
@@ -178,6 +179,7 @@ symmetric-padding=1
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line **NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
``` ```
[property]
... ...
force-implicit-batch-dim=1 force-implicit-batch-dim=1
... ...