Search Results for author:
Found 12 papers, 12 papers with code
Scene Text Recognition Models Explainability Using Local Features
In this work, the problem of interest is Scene Text Recognition (STR) Explainability, using XAI to understand the cause of an STR model's prediction.
EfficientSpeech: An On-Device Text to Speech Model
State of the art (SOTA) neural text to speech (TTS) models can generate natural-sounding synthetic voices.
Scene Text Recognition with Permuted Autoregressive Sequence Models
Context-aware STR methods typically use internal autoregressive (AR) language models (LM).
Ranked #4 on
Scene Text Recognition
on COCO-Text
(using extra training data)
Depth Pruning with Auxiliary Networks for TinyML
Pruning is a neural network optimization technique that sacrifices accuracy in exchange for lower computational requirements.
Improving Model Generalization by Agreement of Learned Representations from Data Augmentation
Experimental results further show that unlike other regularization terms such as label smoothing, AgMax can take advantage of the data augmentation to consistently improve model generalization by a significant margin.
Data Augmentation for Scene Text Recognition
Scene text recognition (STR) is a challenging task in computer vision due to the large number of possible text appearances in natural scenes.
GOO: A Dataset for Gaze Object Prediction in Retail Environments
To this end, we present a challenging new task called gaze object prediction, where the goal is to predict a bounding box for a person's gazed-at object.
Vision Transformer for Fast and Efficient Scene Text Recognition
On a comparable strong baseline method such as TRBA with accuracy of 84. 3%, our small ViTSTR achieves a competitive accuracy of 82. 6% (84. 2% with data augmentation) at 2. 4x speed up, using only 43. 4% of the number of parameters and 42. 2% FLOPS.
Ranked #8 on
Scene Text Recognition
on ICDAR 2003
Next-Best View Policy for 3D Reconstruction
Our experiments show that using Scan-RL, the agent can scan houses with fewer number of steps and a shorter distance compared to our baseline circular path.
Pyramid U-Network for Skeleton Extraction From Shape Points
PSPU-SkelNet is a pyramid of three U-Nets that predicts the skeleton from a given shape point cloud.
A Conditional Generative Adversarial Network for Rendering Point Clouds
In computer graphics, point clouds from laser scanning devices are difficult to render into photo-realistic images due to lack of information they carry about color, normal, lighting, and connection between points.
Fast Disparity Estimation using Dense Networks
Disparity estimation is a difficult problem in stereo vision because the correspondence technique fails in images with textureless and repetitive regions.
