3D Deep Learning with Pytorch3D [Part 1 - Introduction] Setup environment conda create -n dl3d python=3.7 conda activate dl3d conda install pytorch torchvision torchaudio cudatoolkit-11.1 -c p...

Paper note - [Week 3] Monocular 3D Vehicle Detection Using Uncalibrated Traffic Cameras through Homography (IROS 2021) Motivation Traffic cameras are widely deployed today to monitor tr...

Deformable Convolutional Networks 1. Introduction A key issue with traditional convolutional neural networks (CNNs) is their inability to adapt to geometric variations and transformations. Us...

Spatial Transformer Networks Convolutional Neural Networks (CNN) possess the inbuilt property of translation invariance. This enables them to correctly classify an image at test time, even when it...

Spatial Transformer Networks Convolutional Neural Networks (CNN) possess the inbuilt property of translation invariance. This enables them to correctly classify an image at test time, even when it...

Bird Eye View Generation Using IPM (Inverse Perspective Mapping) 1. IPM IPM is short for and its job is to convert Inverse Perspective Mapping 2D images into 3D format. Since it is expressed as ,...

Paper note [Week 2] Simple-BEV: What Really Matters for Multi-Sensor BEV Perception? [ICRA 2023] Motivation Focus on innovating new techniques for “lifting” features from the 2D image plane(...

NeRF - Representing Scenes as Neural Radiance Fields for View Synthesis Review I. Introduction There is a task called view synthesis that takes multiple photos of an object as input and finds o...

[XAI 6] - Human-Friendly Explanations with TCAV In the previous few part, we have extensively discussed LIME and SHAP. You have also seen the practical aspect of applying the Python frameworks of ...

[XAI 5] - Model Interpretability Using SHAP SHapley Additive exPlanation (SHAP) is another popular choice of explainability technique in ML and, in certain scenarios, is more effective than LIME...