yuantangliang / softmaxfocalloss
the loss function in Aritcal ‘Focal Loss for Dense Object Detection‘’
☆17Updated 7 years ago
Related projects: ⓘ
- Diagonalwise Refactorization: An Efficient Training Method for Depthwise Convolutions (in Caffe)☆34Updated 6 years ago
- code fixed from https://github.com/sanghoon/pva-faster-rcnn☆7Updated 7 years ago
- Focal loss for mxnet SSD example.☆23Updated 6 years ago
- A MXNet implementation of Xception☆20Updated 6 years ago
- ☆37Updated 5 years ago
- UnitBox: An Advanced Object Detection Network☆23Updated 7 years ago
- A MXNet implementation of IBN-Net (Resnet-IBN-a, Resnext-IBN-a, Resnext-IBN-b)☆35Updated 5 years ago
- FAIR's research platform for object detection research, implementing popular algorithms like Mask R-CNN and RetinaNet.☆10Updated 6 years ago
- optimized caffe for GBD☆26Updated 7 years ago
- Cascade R-CNN forked from gluon-cv☆33Updated last year
- MXNet port of R-FCN☆51Updated 7 years ago
- unofficial reproduce of detnet☆17Updated 5 years ago
- VGG16 architecture with BatchNorm☆14Updated 7 years ago
- Improving Object Detection from Scratch via Gated Feature Reuse (BMVC 2019)☆65Updated 4 years ago
- ☆24Updated this week
- this repo attemps to reproduce DSOD: Learning Deeply Supervised Object Detectors from Scratch use gluon reimplementation☆14Updated 6 years ago
- Feature Pyramid Networks for Object Detection on mxnet☆31Updated 6 years ago
- Caffe: a fast open framework for deep learning.☆27Updated 7 years ago
- A simplify version of mobilenet, with less group and feature maps, trained on Imagenet.☆17Updated 7 years ago
- the symbol description of mobilenet v2☆11Updated 6 years ago
- The codebase for Zoom Out-and-In Network with Recursive Training for Object Proposal☆22Updated 6 years ago
- ImageNet Detection Tricks☆29Updated 7 years ago
- ShuffleNet implementation in mxnet☆41Updated 6 years ago
- ☆12Updated this week
- FCOS: Fully Convolutional One-Stage Object Detection☆29Updated last year
- ☆12Updated this week
- mxnet deploy version of pseudo-3d-residual-networks(P-3D), sport1m and Kinetics pretrained model is supported☆13Updated 6 years ago
- Top-1 Acc=61.0% on ImageNet, without any sacrificing compared with SqueezeNet v1.1.☆22Updated 7 years ago
- Code for our paper "Generalized Orderless Pooling Performs Implicit Salient Matching" published at ICCV 2017.☆49Updated 6 years ago
- A Gluon implementation of Mnasnet☆59Updated 5 years ago