Comments (6)
Yes, we are planning to release the pre-trained model for WiderPedestrian as well. Stay notified or keep watching the repository. As for CrowdHuman1 and CrowdHuman 2, both are models trained at different point in time. So you might observe a slight performance difference between two models.
from pedestron.
Pre-trained model for wider pedestrian has been released.
from pedestron.
Pre-trained model for wider pedestrian has been released.
Thanks very much. Just tried it, it works the best on my own dataset. But people in my our image is very small like 10*5 pixels, is there a way to increase the input image size so that the network can detect small persons in the test image?
from pedestron.
is there a way to increase the input image size so that the network can detect small persons in the test image?
In the config file you can change the resolution of the image. By the way which dataset are you testing on ?
from pedestron.
is there a way to increase the input image size so that the network can detect small persons in the test image?
In the config file you can change the resolution of the image. By the way which dataset are you testing on ?
Thanks! changed that and get a bit higher accuracy from looking at it. Do you have a plan to update how to train custom dataset using .json as annotation file and also the evaluation code? I am using my own dataset, not one of public datasets listed in your repo. It contains very small persons on the beach, so I am considering retraining from scratch. The pretrained model couldn't handle cases that part of human bodies are hidden by sea. Really appreciate your quick reply.
from pedestron.
Training on a custom dataset should be similar to training on any of the datasets provided in the repo. Create a .json file or your custom dataset and try with the hyper parameters. {perhaps this can be helpful, How to use my own datasets.
from pedestron.
Related Issues (20)
- Docker image is not working! HOT 4
- Impact of mean_teacher on the training process HOT 1
- CSP pretrained weights HOT 1
- Reproduce resutls on Caltech dataset HOT 3
- Information regarding the training HOT 1
- Image_scale of Caltech while training HOT 6
- ImportError: /pedestron/tools/../mmdet/ops/dcn/deform_conv_cuda.cpython-36m-x86_64-linux-gnu.so: undefined symbol: _ZN6caffe26detail37_typeMetaDataInstance_preallocated_32E HOT 1
- RuntimeError: Expected cudaMemcpy(&mask_host[0], mask_dev, sizeof(unsigned long long) * boxes_num * col_blocks, cudaMemcpyDeviceToHost) == cudaSuccess to be true, but got false. (Could this error message be improved? If so, please report an enhancement request to PyTorch.) HOT 1
- Training citypersons with all the instances? HOT 1
- Welcome update to OpenMMLab 2.0
- Runtime error - Training with EuroCity persons HOT 1
- Training Caltech using CSP HOT 1
- What's ignore_other_vru in ECP evaluation? HOT 2
- result has no bbox HOT 5
- mmdet/ops/roi_align/src/roi_align_kernel.cu(145): error: identifier "THCudaCheck" is undefined HOT 1
- Update instruction for 2023 HOT 1
- custom dataset test HOT 1
- Evaluate different pre-trained model
- A request for advice on pedestrian tracking work. HOT 2
- a bug when converting crowdhuman to coco format
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pedestron.