ClothFormer: A Boundary-Aware Self-Attention Network for Human Outfit Parsing

Nov 1, 2023·
Vurimi Bhanu Pranay
,
Nischal DS
,
Bhargav Kumar Nammi
Shiv Kumar
Shiv Kumar
,
Abhilash SK
· 1 min read
Abstract
ClothFormer tackles human outfit parsing with a boundary-aware self-attention network that preserves garment edges while modeling long-range dependencies across body regions and clothing items. The boundary-aware design improves segmentation around fine garment boundaries where standard parsers bleed across adjacent regions. The model uses a self-attention mechanism that attends to clothing regions while suppressing irrelevant background context, achieving improved parsing accuracy on standard benchmarks.
Type
Publication
2023 International Conference on Recent Advances in Information Technology for Sustainable Development (ICRAIS)
Status
Peer-reviewed
publications

Read the paper on IEEE Xplore.

Computer Vision Human Parsing Self-Attention Deep Learning
Shiv Kumar
Authors
Shiv Kumar
Lead Engineer — GenAI / Agentic AI & Applied ML

Lead Engineer with 5 years of hands-on experience developing and deploying GenAI, RAG, Agentic AI, and deep learning systems from scratch. I build production-grade AI platforms for realtime voice interaction, enterprise knowledge retrieval, regulatory search, and automated finance analysis. My work spans NLP, autonomous driving, computer vision, and Camera-LiDAR-Radar fusion, with a focus on reliable architecture, model optimization, and scalable data pipelines.

Stack: Python, GCP, GenAI, LLMs, Agents, Docker, FastAPI, Cloud SQL, Google ADK, Vertex AI, Agent Engine, Cloud Run, OpenAI Realtime API, WebSockets, Redis, MongoDB, Milvus, JWT, Docling, BM25, ModernBERT, vLLM, and FAISS.