PaddleDetection icon indicating copy to clipboard operation
PaddleDetection copied to clipboard

Object Detection toolkit based on PaddlePaddle. It supports object detection, instance segmentation, multiple object tracking and real-time multi-person keypoint detection.

Results 513 PaddleDetection issues
Sort by recently updated
recently updated
newest added

### 问题确认 Search before asking - [X] 我已经查询[历史issue](https://github.com/PaddlePaddle/PaddleDetection/issues),没有报过同样bug。I have searched the [issues](https://github.com/PaddlePaddle/PaddleDetection/issues) and found no similar bug report. ### bug描述 Describe the Bug 找出来的索引越界严重 ### 复现环境 Environment - paddlepaddle-gpu 2.3.1.post116...

### 问题确认 Search before asking - [X] 我已经查询[历史issue](https://github.com/PaddlePaddle/PaddleDetection/issues),没有报过同样bug。I have searched the [issues](https://github.com/PaddlePaddle/PaddleDetection/issues) and found no similar bug report. ### bug描述 Describe the Bug tensorRT 加速失败 如下图 无任何提示就自动结束了程序 ![image](https://user-images.githubusercontent.com/110536580/182605014-0b125220-6196-4d22-a60f-ff6948ac0e05.png) ### 复现环境...

### 问题确认 Search before asking - [X] 我已经搜索过问题,但是没有找到解答。I have searched the question and found no related answer. ### 请提出你的问题 Please ask your question PaddleDetection不支持Voc的instance segmentation数据集吗?

### 问题确认 Search before asking - [x] 我已经搜索过问题,但是没有找到解答。I have searched the question and found no related answer. ### 请提出你的问题 Please ask your question 通过如下命令成功运行对两个demo视频的mtmct: > python deploy/pptracking/python/mot_sde_infer.py --model_dir=output_inference/ppyolov2_r50vd_dcn_365e_aic21mtmct_vehicle/ --reid_model_dir=output_inference/deepsort_pplcnet_vehicle/ --mtmct_dir=mtmct-demo...

### 问题确认 Search before asking - [X] 我已经查询[历史issue](https://github.com/PaddlePaddle/PaddleDetection/issues),没有类似需求。I have searched the [issues](https://github.com/PaddlePaddle/PaddleDetection/issues) and found no similar feature requests. ### 需求描述 Feature Description 对于现实中大图片需要检测的目标不是很多,能不能实现copypaste这个数据增强功能,想着ppocr已经实现了,我看detection还没有实现 ### 是否愿意提交PR Are you willing to submit...

### 问题确认 Search before asking - [X] 我已经搜索过问题,但是没有找到解答。I have searched the question and found no related answer. ### 请提出你的问题 Please ask your question 你好,我在使用solov2导出onnx模型时,出现如下错误。请问是否solov2不支持导出onnx,如果我要在openvino和trt上部署solov2,该使用什么方法呢? ![图片](https://user-images.githubusercontent.com/52130681/182752600-39c90fe0-12c5-4cea-b72d-7603d783b8d8.png)

### 问题确认 Search before asking - [X] 我已经查询[历史issue](https://github.com/PaddlePaddle/PaddleDetection/issues),没有类似需求。I have searched the [issues](https://github.com/PaddlePaddle/PaddleDetection/issues) and found no similar feature requests. ### 需求描述 Feature Description ![QQ图片20220725164101](https://user-images.githubusercontent.com/6490927/180735484-7f4a772d-36aa-4cc7-8347-0b1dcae103c3.jpg) 需要滑窗预测 9000*9000 不能缩小 细节看不到 需要重叠裁剪之后预测 但是预测 这一张...

### 问题确认 Search before asking - [X] 我已经查询[历史issue](https://github.com/PaddlePaddle/PaddleDetection/issues),没有报过同样bug。I have searched the [issues](https://github.com/PaddlePaddle/PaddleDetection/issues) and found no similar bug report. ### bug描述 Describe the Bug PPhuman 使用 --run_mode trt_fp16 参数,程序 运行到Model Configuration...

### 问题确认 Search before asking - [X] 我已经查询[历史issue](https://github.com/PaddlePaddle/PaddleDetection/issues),没有类似需求。I have searched the [issues](https://github.com/PaddlePaddle/PaddleDetection/issues) and found no similar feature requests. ### 需求描述 Feature Description 1. yoloe这种anchor free的方式,针对长宽比很大(比如1:100)的目标检测效果很差。但是工业场景存在大量的长宽比很大的目标。 2. 具体差的原因个人感觉有两个 - 标签分配策略的问题 -...

### 问题确认 Search before asking - [X] 我已经搜索过问题,但是没有找到解答。I have searched the question and found no related answer. ### 请提出你的问题 Please ask your question 我最近学习了ppyoloe主干结构的源码,我发现其主干结构所使用的激活函数和网上画的图不一致,我看代码是,主干结构中除了EffectiveSELayer模块是使用hardsigmoid激活函数以外,其他地方都是使用的swish激活函数。并且我用visualdl中看了网络结构确实是这样的。 但在aistudio加精的项目给出的网络结构图是这样的: ![ppyooe](https://user-images.githubusercontent.com/101441487/182372086-c83dabf7-a7e1-49f7-9ed1-e3866135348e.png) 我看了ppyoloe的论文,里面我好像没有找到比较详细的网络结构,请问你们ppyoloe的主干结构中一直是使用swish激活函数吗?还是之前是使用的relu和silu