- 11
- 1 742
Mehdi Ghasemzadeh
Приєднався 1 лют 2013
Depth Estimation by Camera and LiDAR Fusion (The MVSEC Dataset, Night3)
P: Our network prediction, GT: Ground Truth.
This work proposes a Recurrent-CNN-based network for depth estimation using an event camera, a frame camera, and a LiDAR sensor, we evaluate our model on The MVSEC Dataset.
GitHub page: github.com/mehdighasemzadeh/Depth_Estimation_Camera_LiDAR_Fusion
This work proposes a Recurrent-CNN-based network for depth estimation using an event camera, a frame camera, and a LiDAR sensor, we evaluate our model on The MVSEC Dataset.
GitHub page: github.com/mehdighasemzadeh/Depth_Estimation_Camera_LiDAR_Fusion
Переглядів: 67
Відео
Depth Estimation by Camera and LiDAR Fusion (The MVSEC Dataset, Night2)
Переглядів 507 місяців тому
P: Our network prediction, GT: Ground Truth. This work proposes a Recurrent-CNN-based network for depth estimation using an event camera, a frame camera, and a LiDAR sensor, we evaluate our model on The MVSEC Dataset. GitHub page: github.com/mehdighasemzadeh/Depth_Estimation_Camera_LiDAR_Fusion
Depth Estimation by Camera and LiDAR Fusion (The MVSEC Dataset, Night1)
Переглядів 497 місяців тому
P: Our network prediction, GT: Ground Truth. This work proposes a Recurrent-CNN-based network for depth estimation using an event camera, a frame camera, and a LiDAR sensor, we evaluate our model on The MVSEC Dataset. GitHub page: github.com/mehdighasemzadeh/Depth_Estimation_Camera_LiDAR_Fusion
Depth Estimation by Camera and LiDAR Fusion (The MVSEC Dataset, Day1)
Переглядів 817 місяців тому
P: Our network prediction, GT: Ground Truth. This work proposes a Recurrent-CNN-based network for depth estimation using an event camera, a frame camera, and a LiDAR sensor, we evaluate our model on The MVSEC Dataset. GitHub page: github.com/mehdighasemzadeh/Depth_Estimation_Camera_LiDAR_Fusion
Robustness test of our Event-Frame-Based network
Переглядів 34Рік тому
The robustness test on Event-Scape dataset is done using a wide range of blurred images. A wide range of blurred images are fed to the network, and the network’s predictions are shown in the video, results show our network is robust to the wide degree of blurred images although the accuracy decreases with increasing the blurring, outputs are still reliable. GT: Ground Truth P: Predicted Our Git...
Density-Based Crowd Counting
Переглядів 115Рік тому
This is a video from the results of our network for crowd counting. Density map of people and the number of people are shown in this video. Github: github.com/mehdighasemzadeh/Crowd-Counting-Density-Based
Head Detection using YOLOv5
Переглядів 1,1 тис.Рік тому
Crowd Counting using YOLOv5 on Mall Dataset, the number of people and detected heads are shown in this video. Mall Dataset: personal.ie.cuhk.edu.hk/~ccloy/downloads_mall_dataset.html GitHub: github.com/mehdighasemzadeh/Crowd-Counting-YOLOV5.git
Event-Frame Based Semantic Segmentation, results on Event-Scape dataset
Переглядів 57Рік тому
This is a video that reveals our network performance on Event-Scape dataset. Event-Scape dataset: github.com/uzh-rpg/rpg_ramnet P: Predicted GT: Ground Truth Our GitHub page: github.com/mehdighasemzadeh/Event-Frame-Based-Semantic-Segmentation.git
Event-Frame Based Semantic Segmentation, results on DDD17 dataset
Переглядів 35Рік тому
This is a video from our network performance on DDD17 dataset which is introduced by Ev-SegNet. DDD17 dataset: sensors.ini.uzh.ch/news_page/DDD17.html Ev-SegNet: github.com/Shathe/Ev-SegNet P: Predicted GT: Ground Truth Our GitHub page: github.com/mehdighasemzadeh/Event-Frame-Based-Semantic-Segmentation.git
Event-Based Semantic Segmentation, results on EventScape Dataset
Переглядів 33Рік тому
This is a video from the results of our network on EventScape dataset EventScape dataset: github.com/uzh-rpg/rpg_ramnet P: Predicted GT: Ground Truth Our GitHub page: github.com/mehdighasemzadeh/Event-Based-Semantic-Segmentation.git
Event-Based Semantic Segmentation, results on DDD17 dataset
Переглядів 82Рік тому
This is a video from the results of our network on DDD17 dataset which was introduced by Ev-SegNet DDD17 dataset: sensors.ini.uzh.ch/news_page/DDD17.html Ev-SegNet: github.com/Shathe/Ev-SegNet P: Predicted GT: Ground Truth Our GitHub page: github.com/mehdighasemzadeh/Event-Based-Semantic-Segmentation.git
Can you tell me how to de-identify the face using the red circle??
After the head detection, the red circles are put on the detected heads. Opencv is used for that.
@@mehdighasemzadeh what you use for training data??
@@현김우-p8dFDDB: paperswithcode.com/dataset/fddb and head detection dataset: github.com/HCIILAB/SCUT-HEAD-Dataset-Release were used for training.
@@mehdighasemzadeh thanks have a nice day😊