Focused on computer vision, image processing, color science, and AI algorithms; sharing hardcore tech, engineering practice, and deep insights as a diligent AI technology practitioner.
The ICCV MIPI workshop introduces the ViDA-UGC competition, presenting a richly annotated UGC image quality dataset, a benchmark suite covering degradation detection, region perception, and quality description, detailed evaluation metrics, submission formats, prize information, and open participation for researchers worldwide.
FilMaster is a pioneering AI system that learns cinematic principles from a 440,000‑shot movie database, combines multimodal LLMs, RAG, and audience‑centric rhythm control to generate editable, high‑quality films, and outperforms prior methods by over 50% on the new FilmEval benchmark.
The paper introduces Mamba-Adaptor, a plug‑and‑play module combining Adaptor‑T and Adaptor‑S to overcome causal computation, long‑range forgetting, and spatial modeling limits of visual Mamba models, delivering top‑ranked results on ImageNet and COCO across multiple downstream tasks.
The article surveys 117 recent multimodal‑fusion papers, classifies them into improvement‑based and combination‑based approaches, highlights representative works such as TimeXL, OGP‑Net, MMR‑Mamba and FusionSight, and provides a free collection of papers, classic models and code repositories for researchers.
Large language models are booming, and this article highlights six cutting‑edge research directions—LLM‑plus synthetic data, reward modeling, inference techniques, LLM‑as‑a‑Judge, safety alignment, and long‑context handling—each illustrated with recent papers, experimental results, and links to code repositories.
Nvidia introduces SeNaTra, a native‑segmentation vision transformer that replaces uniform down‑sampling with a content‑aware spatial grouping layer, delivering superior zero‑shot and supervised segmentation performance while cutting parameters and FLOPs compared with Swin Transformer and other backbones.
DeepKD introduces a double‑layer decoupling framework and a dynamic top‑K mask that adaptively denoises low‑confidence logits, addressing conflicts between target and non‑target knowledge flows; extensive experiments on CIFAR‑100, ImageNet‑1K, and MS‑COCO demonstrate consistent accuracy gains and state‑of‑the‑art performance.
The NTIRE 2025 challenge introduced a new benchmark for user‑generated content video enhancement, detailing a 150‑video dataset, a pairwise subjective evaluation using the Bradley‑Terry model, hardware specifications, and the diverse multi‑stage deep‑learning methods and results of participating teams.
The paper presents a tree‑guided convolutional neural network that leverages binary‑tree structures, cosine‑based cross‑domain feature extraction, and an adaptive Nesterov momentum optimizer to enhance key layer interactions, achieving superior image super‑resolution performance as demonstrated by extensive experiments.
ImgEdit introduces a large‑scale, high‑quality editing dataset and the ImgEdit‑Bench benchmark, detailing a robust data‑generation pipeline, multi‑round editing tasks, and a specialized evaluation model, and demonstrates through extensive experiments that its ImgEdit‑E1 model outperforms existing open‑source editors and narrows the gap with closed‑source systems.
