Computer Vision

DaTaobao Tech

Mar 6, 2024 · Artificial Intelligence

AI Clothing Graffiti Project: Implementation and Optimization of AIGC Technology in Taobao Life 2

The AI Clothing Graffiti Project in Taobao Life 2 leverages Stable Diffusion, ControlNet, and LoRA to let users generate and stylize clothing designs via text‑image prompts, employing parallel processing, face repair, and content filtering, and has launched successfully, inviting algorithm engineers to join the team.

AIGCComputer VisionControlNet

0 likes · 14 min read

AI Clothing Graffiti Project: Implementation and Optimization of AIGC Technology in Taobao Life 2

Alimama Tech

Jun 14, 2023 · Artificial Intelligence

Intelligent Live‑Streaming Video Editing Techniques and Practices

Alibaba Mama’s end‑to‑end intelligent clipping system automatically transforms long live‑stream e‑commerce videos into short, high‑quality ads by segmenting streams, classifying speech with GPT‑based tags, selecting visually appealing clips, arranging coherent storylines, and applying effects, achieving 96% classification accuracy and improved advertising efficiency.

Computer VisionMultimodalVideo Editing

0 likes · 14 min read

Intelligent Live‑Streaming Video Editing Techniques and Practices

Alimama Tech

Jul 13, 2022 · Artificial Intelligence

Fully Automatic Template‑Free Image‑Text Creative Generation System

Alibaba Alimama’s fully automatic, template‑free image‑text creative generation system uses deep‑learning models across material mining, layout synthesis, on‑image copy generation, and visual attribute rendering to produce personalized ad creatives directly from product images and metadata, achieving roughly 19 % CTR lift over prior template‑based methods.

Computer VisionGenerative ModelsLayout Generation

0 likes · 19 min read

Fully Automatic Template‑Free Image‑Text Creative Generation System

DaTaobao Tech

May 24, 2022 · Artificial Intelligence

GEN-VLKT: Simplify Association and Enhance Interaction Understanding for HOI Detection

GEN‑VLKT introduces a Guided‑Embedding Network with position‑ and instance‑guided embeddings to remove costly post‑processing and leverages CLIP‑based visual‑linguistic knowledge transfer for interaction understanding, achieving state‑of‑the‑art HOI detection performance and zero‑shot capability, now deployed in Alibaba’s Taobao services.

ClipComputer VisionHOI detection

0 likes · 7 min read

GEN-VLKT: Simplify Association and Enhance Interaction Understanding for HOI Detection

Xiaohongshu Tech REDtech

Apr 26, 2022 · Artificial Intelligence

Multi-Modal Technology in Intelligent Creation: Insights from REDtech Live

At REDtech Live, leading researchers showcased advances in multi‑modal technology for intelligent creation, covering efficient video‑text retrieval, semantic voice synthesis, low‑cost neural 3D reconstruction, deep‑learning‑driven visual content generation, and pixel‑level video segmentation with 2D‑3D fusion techniques.

3D DigitalizationAI applicationsComputer Vision

0 likes · 7 min read

Multi-Modal Technology in Intelligent Creation: Insights from REDtech Live

DeWu Technology

Nov 18, 2020 · Artificial Intelligence

AR Fundamentals and Shoe Try‑On Implementation

The presentation explains AR fundamentals, distinguishes it from AI and VR, and details a shoe‑try‑on system that captures 30 fps video, uses AI key‑point detection and pose estimation to overlay 3D shoe models—created via manual, scanning, or photogrammetry methods—rendered with GPU pipelines and PBR, enhanced by green‑screen occlusion and shadow techniques, earning positive audience feedback.

3D modelingARComputer Vision

0 likes · 7 min read

AR Fundamentals and Shoe Try‑On Implementation

iQIYI Technical Product Team

Oct 16, 2020 · Artificial Intelligence

Cartoon Face Recognition: Introducing the iCartoonFace Benchmark Dataset

iQIYI’s ACM Multimedia‑accepted paper unveils iCartoonFace, the world’s largest manually annotated cartoon‑face dataset—over 5,000 characters and 400,000 real‑scene images—accompanied by a semi‑automatic collection pipeline and multi‑person training framework, now powering AI services, large‑scale contests and accelerating cartoon‑character recognition research.

Cartoon Face RecognitionComputer VisionDataset

0 likes · 4 min read

Cartoon Face Recognition: Introducing the iCartoonFace Benchmark Dataset

iQIYI Technical Product Team

Jul 26, 2019 · Artificial Intelligence

Preface

In the 2019 iQIYI Celebrity Video Identification Challenge, our team secured fifth place by accurately recognizing video identities using mAP scoring, and this article shares the strategies, insights, and experiences of the top‑five teams, emphasizing a straightforward, pragmatic approach championed by iQIYI’s technology product team.

ChallengeComputer VisionTechnical Report

0 likes · 5 min read

iQIYI Technical Product Team

Jul 5, 2019 · Artificial Intelligence

iQIYI Multimodal Person Recognition Competition: 91.14% Accuracy Achieved by BUPT Team

After a three‑month contest co‑hosted by iQIYI and ACM MM, 255 teams competed on the challenging iQIYI‑VID‑2019 multimodal dataset, and the BUPT Automation School team won with a 91.14% person‑recognition accuracy, advancing the field and enhancing iQIYI’s video recommendation and AI services.

AI competitionComputer VisionDeep Learning

0 likes · 6 min read

iQIYI Multimodal Person Recognition Competition: 91.14% Accuracy Achieved by BUPT Team

Tencent Cloud Developer

Aug 6, 2018 · Artificial Intelligence

Tencent's AI Breast Cancer Screening System: Technical Architecture and Implementation

Tencent's AI Breast System combines mammography, pathology, MRI and ultrasound analysis using a multi‑scale, progressive TMuNet model that processes four views, learns from physician feedback, and delivers lesion localization, malignancy scoring and automated reports, achieving up to 92% sensitivity and reducing annotation time.

AI Medical ImagingBreast Cancer DetectionComputer Vision

0 likes · 13 min read

Tencent's AI Breast Cancer Screening System: Technical Architecture and Implementation