398. Findings Report
▮ Sharing your findings In many cases, at the final phase of your data science project, you will need to organize and represent your findings to your audiences through a report as a deliverable. These reports can be in many…
Category Research Paper
359. Future Frame Prediction For Anomaly Detection
Anomaly Detection Anomaly detection in videos refers to the identification of events that do not align with the expected behavior. This paper is a pioneering work that leverages the difference between a predicted future frame and its ground truth to…
358. Different Settings of Transfer Learning
Different Approaches Here are the differences between the concepts of transfer learning depending on which types of data are available when training. Reference A Survey on Transfer Learning
354. ConvNeXt
“Modernizing” ConvNets “ConvNext” is a gradually “modernized” traditional ConvNet model designed to reexamine the design spaces and test the limits of what a pure ConvNet can achieve. The paper does this by modifying the Micro Design of the ConvNet architecture…
353. Normalization Methods
Batch Normalization Batch Normalization is a milestone technique in the development of deep learning, enabling various networks to train. However, BN’s error increases rapidly when the batch size becomes small affecting the batch statistics estimation. Furthermore, the concept of “batch”…
352. Swin Transformer
Abstract In existing Transformer-based models, tokens are all of a fixed scale, a property unsuitable for vision applications such as semantic segmentation that require dense prediction at the pixel level. In addition, due to the computational complexity of its self-attention…
351. Gaussian Error Unit (GELU)
Structure Gaussian Error Unit is a high-performing neural network activation function that weights inputs by their value, rather than gates inputs by their sign as in ReLUs. GELU is defined as the equation in the image. Results GELU exceeds the…
347. Point Pillar: 3D Object Detection from Point Clouds
Abstract Point Pillar is an architecture proposed for 3D object detection using point clouds as inputs. Architecture The architecture consists of mainly 3 elements. Pillar Feature Net BackBone Detection Head 1. Pillar Feature Net This phase takes the following steps.…
346. CaDDN
Depth Estimation The main challenges in monocular 3D object detection lie in accurately predicting object depth. CaDDN(Categorical Depth Distribution Network) uses a predicted categorical depth distribution for each pixel to project appropriate depth in 3D space. Approaches There are several…
343. Depth From Motion
Types Here are some approaches to measure depth by only using RGB images from motion. Optical Expansion Observe how the length of an object changes as the camera moves closer. If the object is close, the length will dramatically change…