-
š Hi, Iām ParitoshParmar
-
š Iām interested in Computer Vision, especially, Video Understanding
-
āØ My latest projects:
- Hierarchical NeuroSymbolic Approach for Comprehensive and Explainable Action Quality Assessment [š CVPR 2024 CVSports Best Paper Award (Oral Presentation; 3/3 Strong Accepts); Selected for CVPR 2024 Demos]: Not just a single score prediction, but Detailed, Unbiased Action Quality Assessment Report Generation, complete with Visual Evidence!
- Fine-grained Exercise Action Quality Assessment: Novel Domain Knowledge-Informed Self-Supervised Pose-Motion Contrastive Approaches for Fine-grained Action Quality Assessment (can be used for Diving as well!) + Fitness-AQA dataset
- Multimodal Piano Skills Assessment: Audio CNN + Video CNN + Multimodal CNN Approaches for Piano Skills Assessment + Multimodal Dataset for Piano Skills Assessment
- Novel Test Benchmark for Truer Video Understanding: Novel Challenging Video Understanding Dataset -- Win-Fail Action Understanding dataset + In-Depth Analysis of various Video Understanding baseline models
- Multitask Action Quality Assessment: Levaraging Fine-grained, Detailed Action Description Generation and Fine-grained Action Recognition to Improve Performance on Action Quality Assessment + Better Generalization from Diving to Gymnastic Vault + Largest Multilabel Diving AQA Dataset
paritoshparmar Goto Github PK
Name: Paritosh Parmar
Type: User
Company: University of British Columbia
Bio: Computer Vision Researcher