GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → ishmav16 → Dual-CNN-Models-for-Unsupervised-Monocular-Depth-Estimation

ishmav16 / Dual-CNN-Models-for-Unsupervised-Monocular-Depth-Estimation

Licence: other
Dual CNN Models for Unsupervised Monocular Depth Estimation

Programming Languages

python
139335 projects - #7 most used programming language
shell
77523 projects

Labels

tensorflowcnnautoencoderunsupervised-learningdepth-estimation

Projects that are alternatives of or similar to Dual-CNN-Models-for-Unsupervised-Monocular-Depth-Estimation

learning-topology-synthetic-data
Tensorflow implementation of Learning Topology from Synthetic Data for Unsupervised Depth Completion (RAL 2021 & ICRA 2021)
Stars: ✭ 22 (-38.89%)
Mutual labels:  unsupervised-learning, depth-estimation
SESF-Fuse
SESF-Fuse: An Unsupervised Deep Model for Multi-Focus Image Fusion
Stars: ✭ 47 (+30.56%)
Mutual labels:  autoencoder, unsupervised-learning
Indoor-SfMLearner
[ECCV'20] Patch-match and Plane-regularization for Unsupervised Indoor Depth Estimation
Stars: ✭ 115 (+219.44%)
Mutual labels:  unsupervised-learning, depth-estimation
2D-and-3D-Deep-Autoencoder
Convolutional AutoEncoder application on MRI images
Stars: ✭ 57 (+58.33%)
Mutual labels:  autoencoder, unsupervised-learning
Calc
Convolutional Autoencoder for Loop Closure
Stars: ✭ 119 (+230.56%)
Mutual labels:  autoencoder, unsupervised-learning
Splitbrainauto
Split-Brain Autoencoders: Unsupervised Learning by Cross-Channel Prediction. In CVPR, 2017.
Stars: ✭ 137 (+280.56%)
Mutual labels:  autoencoder, unsupervised-learning
Self-Supervised-depth
Self-Supervised depth kalilia
Stars: ✭ 20 (-44.44%)
Mutual labels:  unsupervised-learning, depth-estimation
Pyod
A Python Toolbox for Scalable Outlier Detection (Anomaly Detection)
Stars: ✭ 5,083 (+14019.44%)
Mutual labels:  autoencoder, unsupervised-learning
Image similarity
PyTorch Blog Post On Image Similarity Search
Stars: ✭ 80 (+122.22%)
Mutual labels:  autoencoder, unsupervised-learning
Tybalt
Training and evaluating a variational autoencoder for pan-cancer gene expression data
Stars: ✭ 126 (+250%)
Mutual labels:  autoencoder, unsupervised-learning
Danmf
A sparsity aware implementation of "Deep Autoencoder-like Nonnegative Matrix Factorization for Community Detection" (CIKM 2018).
Stars: ✭ 161 (+347.22%)
Mutual labels:  autoencoder, unsupervised-learning
Face-Landmarking
Real time face landmarking using decision trees and NN autoencoders
Stars: ✭ 73 (+102.78%)
Mutual labels:  autoencoder
dltf
Hands-on in-person workshop for Deep Learning with TensorFlow
Stars: ✭ 14 (-61.11%)
Mutual labels:  autoencoder
Clustering Algorithms from Scratch
Implementing Clustering Algorithms from scratch in MATLAB and Python
Stars: ✭ 170 (+372.22%)
Mutual labels:  unsupervised-learning
unsup temp embed
Unsupervised learning of action classes with continuous temporal embedding (CVPR'19)
Stars: ✭ 62 (+72.22%)
Mutual labels:  unsupervised-learning
GATE
The implementation of "Gated Attentive-Autoencoder for Content-Aware Recommendation"
Stars: ✭ 65 (+80.56%)
Mutual labels:  autoencoder
lxa5
Linguistica 5: Unsupervised Learning of Linguistic Structure
Stars: ✭ 27 (-25%)
Mutual labels:  unsupervised-learning
salt iccv2017
SALT (iccv2017) based Video Denoising Codes, Matlab implementation
Stars: ✭ 26 (-27.78%)
Mutual labels:  unsupervised-learning
Image-Retrieval
Image retrieval program made in Tensorflow supporting VGG16, VGG19, InceptionV3 and InceptionV4 pretrained networks and own trained Convolutional autoencoder.
Stars: ✭ 56 (+55.56%)
Mutual labels:  autoencoder
tensorflow-mnist-AAE
Tensorflow implementation of adversarial auto-encoder for MNIST
Stars: ✭ 86 (+138.89%)
Mutual labels:  autoencoder
View All Similar Projects ➔

Dual CNN Approach For Depth Estimation

Abstract: The objective of this work is to estimate the depth map that has the information about how far the camera captured the scene. A lot of progress has been made to solve the depth estimation problem in stereo vision. Though, a very satisfactory performance is observed by utilizing the deep learning in supervised manner for depth estimation. This approach needs huge amount of ground truth training data as well as depth maps which is very laborious to prepare and many times it is not available in real scenario. Thus, the unsupervised depth estimation is the recent trend by utilizing the binocular stereo images to get rid of depth map ground truth. In unsupervised depth computation, the disparity images are generated by training the CNN with an image reconstruction loss based on the epipolar geometry constraints. The effective way of using CNN as well as investigating the better losses for the said problem needs to be addressed. In this, auto encoder algorithm based dual CNN based model is presented for unsupervised depth estimation with 6 losses (DNM6) with individual CNN for each view to generate the corresponding disparity map. The proposed dual CNN model is also extended with 12 losses (DNM12) by utilizing the cross disparities. The presented DNM6 and DNM12 models are tested over KITTI driving database and compared with the recent state-of-the-art result of unsupervised depth estimation.

Sample Input Image :

input-image

Depth Image using DNM6 Model :

output-depth-map

Depth Image using DNM12 Model :

output-depth-map

For more details, you can look into the paper here.

Requirements :

Tensorflow 3.0 CUDA 8.0 and Ubuntu

This network is implemented in TensorFlow contains 62 million trainable parameters which include weights and biases and trained in NVIDIA Tesla K80 24GB GDDR5 GPU server up to 50 epochs which takes 49 hours.

Proposed Models :

Developed the 2 models involving 2 CNNs named 'Dual CNN' for estimating the depth map.

Model-1: Dual Network Model with 6 losses (DNM6)

first-approach

Model-2: Dual Network Model with 12 losses (DNM12)

second-approach

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].