The AT‑ADD competition, organized for ACM MM 2026, invites researchers to develop robust audio deepfake detection models across speech, environmental sounds, singing, and music, providing diverse real‑world datasets, baseline code, clear evaluation metrics, and a two‑stage submission process to advance AI security.
This article recounts how India’s rise in software outsourcing inspired China, highlights two Indian inventions—USB and the Discrete Cosine Transform (DCT)—and explains how DCT’s breakthrough image compression paved the way for JPEG, MP3, and modern multimedia streaming.
H.266/VVC, the next‑generation video coding standard finalized in 2020, delivers roughly 50 % bitrate savings over H.265/HEVC with modestly higher decoding complexity, introduces advanced intra‑ and inter‑prediction, transform, quantization, entropy and loop‑filtering tools, and faces patent‑pool and adoption challenges before widespread smartphone integration around 2026.
At the China Multimedia 2024 conference in Yinchuan, Kuaishou presented its latest AI‑driven large‑model technologies—including text‑to‑image, text‑to‑video, and audio models—alongside advances in intelligent video coding, a new research‑fund initiative, and recent industry awards.
Bilibili’s Multimedia Lab created an end‑to‑end intelligent restoration system that assesses video resolution, frame‑rate and quality, then automatically selects and applies image‑level enhancement, frame‑rate up‑sampling, background and face restoration, and optical‑flow interpolation to transform blurry, jittery, artifact‑laden legacy videos into clear, smooth, high‑quality streams, now deployed for on‑demand content and slated for live‑stream expansion.
FFmpeg provides a comprehensive, cross‑platform hardware acceleration framework that abstracts diverse GPU and dedicated video codec interfaces, defines HWContext types, device and frame contexts, and various codec configuration methods, enabling efficient video encoding, decoding, and filtering while addressing performance, compatibility, and pipeline complexity challenges.
The article curates standout AI papers from 2023 CCF‑A conferences—including CVPR, ICLR, ACM MM, and INFORMS—showcasing advances such as Swin‑Transformer video quality assessment, cross‑modal e‑commerce product search, transformer‑based vehicle routing heuristics, diffusion‑driven dance generation, and reinforcement‑learning inventory replenishment.
The Haro Street Cat mobile team created a custom multimedia framework that wraps FFmpeg 4.2.2 in a C++ core library with Android/iOS compatibility layers and Java wrappers for transcoding, live streaming, and composition, delivering hardware‑accelerated decoding, flexible filter pipelines, and reliable transcoding that boosted coverage to over 99 %, cut storage by more than 30 %, accelerated video start‑up, and improved streaming and watermarking performance.
Bilibili’s AI‑driven frame‑interpolation pipeline upgrades low‑frame-rate videos to smooth high‑frame-rate 1080p playback by optimizing optical‑flow models for large motion, texture and text artifacts, pruning for speed, and deploying via the BVT SDK across on‑demand and live streams.
This article introduces Python's multimedia service modules, explaining how to process raw audio data, read and write various audio file formats, detect image and sound file types, convert color systems, and access OSS‑compatible audio devices, all illustrated with practical code examples.
This article presents Weibo's comprehensive multimedia content understanding pipeline, covering video summarization techniques, quality assessment models, OCR advancements, video embedding strategies, and the performance benefits of CV‑CUDA acceleration, while highlighting real‑world applications and engineering trade‑offs.
Alibaba’s Taobao VQA team won the CVPR NTIRE 2023 Video Enhancement Challenge by introducing MD‑VQA, a multi‑dimensional no‑reference video quality model that combines a Swin‑Transformer‑V2 spatial backbone, a pre‑trained SlowFast motion encoder, and a convolutional fusion module, pre‑trained on LSVQ, fine‑tuned on NTIRE data, and augmented spatio‑temporally, achieving state‑of‑the‑art SROCC and PLCC scores and now powering quality monitoring on Alibaba’s live‑streaming and short‑video services.
The article defines pipeline domain design concepts, compares major multimedia frameworks such as FFmpeg, GStreamer, MediaPipe and AVPipeline, and demonstrates a configurable, extensible node‑based architecture that enables fast plugin integration and adaptable audio‑video pipelines across diverse business scenarios and platforms.
FFmpeg 6.0 “Von Neumann” has been officially released, introducing numerous new encoders, decoders, and filters, adding ABI versioning to major releases, deprecating old APIs, and enhancing CLI performance with threading, statistics options, and file‑based filter options, while outlining upcoming features for version 6.1.
FFmpeg 6.0 "Von Neumann" introduces a host of new codecs, decoders, filters, CLI enhancements, ABI versioning, and a more frequent release cadence, offering developers expanded multimedia processing capabilities and improved performance across platforms.
FFmpeg 6.0 "Von Neumann" introduces a host of new decoders, encoders, filters, hardware‑accelerated AV1 support, ABI versioning, and numerous performance and API improvements, marking a major, more structured release cycle for the multimedia framework.
The article provides a comprehensive overview of digital watermarking, covering its fundamental concepts, security features, embedding/detection/extraction processes, major algorithm families, practical applications such as copyright protection and anti‑counterfeiting, and future research directions in multimedia information security.
This article explains how Tencent's advertising system leverages multimedia AI techniques—including multi‑armed bandit, pairwise learning, and DeepFM models—to automatically select optimal templates, music, and stickers for image and video assets, thereby reducing production cost, improving creative quality, and boosting ad performance.
This article explains how to use Three.js to create a realistic 3D phone product page by loading and applying multimedia assets such as custom fonts, textures, audio sources, and video textures, and demonstrates interactive features like ray‑casting for material switching and first‑person controls.
This article explains the OpenMAX (OMX) multimedia acceleration standard, its three-layer architecture (Application, Integration, Development), how it’s used in frameworks like gStreamer, StageFright, and the AVProcessEngine plugin library to modularize audio, video, and image algorithms across Android, iOS, macOS, and Windows platforms.
The article introduces DeViT, a novel Deformed Vision Transformer framework for video inpainting that leverages a deformable patch homography estimator, mask‑pruned attention, and spatio‑temporal weight adaptation, achieving state‑of‑the‑art results on benchmark datasets and highlighting its potential for advanced video editing tools.
Researchers from Kuaishou and Xi’an University of Electronic Science and Technology presented a novel deep‑learning‑based SDR‑to‑HDR video conversion method at ACM Multimedia 2022, introducing hierarchical dynamic context feature mapping, a layered dynamic feature modulation module, and a patch‑discriminator GAN that together achieve superior objective and subjective HDR quality.
At ACM MM 2022, Youku Moku Lab chaired the pioneering QoEVMA workshop, receiving 14 international submissions and accepting eight papers that advanced Quality of Experience assessment for diverse visual multimedia formats, while showcasing the lab’s free‑viewpoint video technology deployed in major events such as the 2022 Beijing Winter Olympics.
The article walks through ffplay’s end‑to‑end playback pipeline—starting with protocol and container demuxing, initializing FFmpeg and SDL, spawning read and decoder threads, handling video/audio decoding, synchronizing streams, and finally rendering frames—offering design insights for constructing a basic media player.
Bilibili’s WasmPlayer is a WebAssembly‑based HEVC software decoder that runs entirely in C/C++, handling demuxing, decoding, resampling and A/V sync on worker threads, using SIMD, AudioWorklet and OffscreenCanvas to deliver smooth 4K 60 fps web video playback with reduced bandwidth and CPU load.
MoviePy is an open‑source Python library licensed under MIT that enables cross‑platform video editing tasks such as cutting, concatenating, adding titles, compositing, and processing audio‑video files—including GIFs—through a simple API demonstrated with a complete code example.
This article presents Tencent's advertising multimedia AI platform, detailing its intelligent video creation engine, fine‑grained ad content understanding, large‑scale similar‑ad retrieval system, and automated ad review pipeline, while also introducing the team and current recruitment opportunities.
This comprehensive guide introduces multimedia front‑end development, explains W3C media standards and HTML elements, explores media APIs, outlines playback scenarios and solutions, and details both consumer‑facing live video systems and production‑side tools such as streaming and video‑editing, while sharing Alibaba’s roadmap for the field.
Celebrating two decades of the open‑source FFmpeg project, this article explains how its multimedia decoding libraries underpin popular video players and major platforms, recounts the 2011 split that birthed Libav, discusses licensing obligations under LGPL/GPL, and reveals the infamous “shame pillar” list of software that ignored those rules.
In this candid talk, a former Alibaba multimedia front‑end engineer shares his career journey—from early patents and B2B animation work to leading live‑streaming projects, detailing the challenges, technical breakthroughs, and personal reflections that helped him overcome professional uncertainty.
Lin Wan shares her career path from a fresh graduate at Alibaba International Station to a multimedia front‑end leader in Taobao Live, detailing personal growth, patent achievements, technical breakthroughs, and lessons on overcoming front‑end career confusion.
This article explains how Alibaba’s multimedia front‑end team built a same‑layer rendering video player for H5, detailing the architecture, performance advantages, degradation strategies, issues encountered during the 618 promotion, and future plans to improve playback experience across live and on‑demand scenarios.
AV1, the royalty‑free video codec created by the Alliance for Open Media and supported by Tencent’s Multimedia Lab, combines advanced block partitioning, intra‑ and inter‑prediction, versatile transforms, sophisticated entropy coding, and loop‑filtering to cut bitrates by roughly 20% for streaming, cloud transcoding, 8K, VR, and medical imaging, while paving the way toward future AV2 and VVC standards.
FFmpeg 4.3 has been released, adding support for TrueHD in MP4, Intel QSV‑accelerated MJPEG and VP9 decoding, Vulkan‑based AMD AMF encoding on Linux, AV1 encoding via rav1e, ZeroMQ, VDPAU VP9 decoding, and numerous filter enhancements.
This article explores the front‑end multimedia technologies that power Taobao Live, covering video and audio fundamentals, container and codec formats, streaming protocols, player architecture, web media APIs, and popular open‑source frameworks for building robust live‑streaming experiences.
iQIYI has become the first Chinese video platform to deploy the open‑source AV1 codec on its PC web and Android apps, cutting video file size by about 20%, and its in‑house QAV1 encoder further reduces bitrate over 40% and runs roughly five times faster than competing encoders, enabling smoother 4K/8K streaming while saving bandwidth.
This article traces the evolution of call center technology across four generations—from early hotlines using PSTN and PBX, through IVR and CTI innovations, to modern multimedia channels—highlighting key concepts, features, and their impact on operational efficiency and customer service.
The CBVRP (Content-Based Video Relevance Prediction) challenge, co‑hosted by Hulu and ACM MM 2019, showcased the shift from user‑based collaborative filtering to content‑driven recommendation, highlighted winning teams and their papers, and underscored the ongoing research importance of cold‑start video recommendation for streaming platforms.
With 5G accelerating video traffic, immersive media relies on advanced 3D representation, efficient point‑cloud and video‑based compression, standardized containers, and selective streaming to enable realistic VR/AR experiences across gaming, autonomous driving, and broadcasting, while Tencent’s lab drives standards and future XR applications.
Discover a curated list of ten indispensable, free, open‑source Linux applications—from package managers and media players to office suites and security tools—each with key features, usage tips, and direct download links to enhance your Linux desktop experience.
The article outlines the technical challenges of short‑video apps and presents the modular, extensible architecture of the 58 Short Video SDK, detailing its layered design, design principles, advantages, and future evolution to support advanced features such as AR, hardware decoding, and h265 encoding.
During the CCF Multimedia Committee’s visit to iQIYI, senior researchers and professors presented cutting‑edge AI, knowledge‑graph‑driven content distribution, image‑text sentiment matching, and intelligent multimedia transmission technologies, while interactive tours of studios and labs deepened academia‑industry collaboration and highlighted iQIYI’s innovative multimedia ecosystem.
This article introduces FFmpeg, outlines its eight core libraries, describes the main command‑line tools (ffmpeg, ffplay, ffprobe), and provides a step‑by‑step example of converting an MP4 video to HEVC with MP3 audio on Windows, including useful help commands and additional features.
Three Alibaba iDST papers on the AI‑driven City Brain were selected for ACM Multimedia 2017, showcasing novel video anomaly detection, deep siamese re‑identification, and stylized image generation methods that improve urban traffic management and demonstrate the broader potential of Alibaba’s ET Brain platform.
