antran89 / my-very-deep-caffe
This is an implementation of very deep two stream CNNs for action recognition. The implementation is inspired by Wang et. al., https://github.com/yjxiong/caffe. Some improvements from Wang's implementation include reading videos from LDMB database, faster testing using LDMB interface. The aim is to work better with big dataset such as UCF101, HM…
☆21Updated 6 years ago
Related projects ⓘ
Alternatives and complementary repositories for my-very-deep-caffe
- The repository to contain codes and models for paper "Two-stream Flow-guided Convolutional Attention Networks for Action Recognition".☆22Updated 6 years ago
- Comparisons between available optical flow implementations using MATLAB.☆32Updated 9 years ago
- This is a BoW implementation for action recognition problem in Matlab.☆11Updated 6 years ago
- Dual Deep Network for Visual Tracking☆14Updated 7 years ago
- Learning Detection with Diverse Proposals at CVPR 2017☆31Updated 7 years ago
- Object detection using AZ-Net☆44Updated 7 years ago
- Attend Refine Repeat: Active Box Proposal Generation via In-Out Localization☆62Updated 5 years ago
- ☆21Updated 7 years ago
- Spatial Transformer Networks☆76Updated 8 years ago
- a branc caffe with feature of Data Augmentation using a configurable stochastic combination of 7 data augmentation techniques.☆32Updated 9 years ago
- Actionness Estimation Using Hybrid Fully Convolutional Networks☆30Updated 8 years ago
- Project Page☆35Updated 8 years ago
- UCF Sports annotations: This repository provides human bounding box annotations of UCF Sports dataset and a function to read these annota…☆14Updated 9 years ago
- code for DeepProposal paper presented in ICCV 2015☆35Updated 8 years ago
- Improved trajectories for action recognition☆22Updated 8 years ago
- Code for "Predictive-Corrective Networks for Action Detection"☆16Updated 6 years ago
- 2016 ActivityNet action recognition challenge. CNN + LSTM approach. Multi-threaded loading.☆54Updated 8 years ago
- Trajectory-pooled Deep-Convolutional Descriptors☆104Updated 7 years ago
- code for Two-Stream SR-CNNs for Action Recognition in Videos 2016☆34Updated 6 years ago
- Implement of 'Beyond Frontal Faces: Improving Person Recognition Using Multiple Cues' by Ning Zhang, et al☆24Updated 9 years ago
- Weakly Supervised Object Localization with Progressive Domain Adaptation (CVPR 2016)☆62Updated 8 years ago
- Repository for Neural Tensor Fusion Network☆20Updated 7 years ago
- Active Convolution☆60Updated 7 years ago
- FBN: Factorized Bilinear Models for Image Recognition (ICCV 2017)☆68Updated 6 years ago
- Master Project @ Hasso-Plattner-Institute. Video classification with deep convolutional networks☆25Updated 9 years ago
- source code for the paper "Hard-Aware-Deeply-Cascaed-Embedding"☆32Updated 7 years ago
- Fine-tuning an already learned model, adapts the architecture to other datasets☆28Updated 7 years ago