arXiv subjects

Xiaoqian Zhu

Publications and source records attributed to Xiaoqian Zhu.

3 recordsLinked to original sources

arXiv2026

FilmBench: A Film-Grade Benchmark for Cinematic Video Generation

Progress in video generation keeps narrowing the visual gap between AI-generated and professionally produced footage, yet most benchmarks still draw prompts from web sources or LLM templates and score them with untrained, generic multimodal models. More fundamentally, their evaluation taxonomies remain rudimentary (overall visual quality, coarse text alignment and temporal smoothness) rather than the professional Cinematic Language criteria by which films are actually made and judged, so they assess basic video plausibility rather than film-grade craft. We introduce FilmBench, a text-to-video (T2V) and reference-to-video (R2V) benchmark grounded in the professional Cinematic Language of the film- academy tradition and co-developed with directors and faculty from the Beijing Film Academy and the Hujing Digital Media & Entertainment Group film studio. It rests on three choices. First, prompts are reverse-engineered from clips of award-winning films spanning 20 cinematic genres and chosen by professional directors, so every prompt is anchored to a verified live-action reference; the prompts follow real shot lists, and most script multiple shots (1,056 of the 1,169 prompts are multi-shot), unlike prior single-clip benchmarks. Second, evaluation follows a three-level Cinematic taxonomy of 3 axes, 12 components and 35 (T2V) +3 (R2V-only) sub-metrics. Third, we develop an in-house expert-grade automatic evaluation agent and open-source its core suite of Cinematic Language operators (FilmOps). Benchmarking leading video generation models (9 for T2V, 7 for R2V), the evaluator reproduces the human model ranking at model-level Spearman \r{ho} = 0.95 (T2V) and 0.96 (R2V). Scores fall well below prior web-style benchmarks, with two consistent gaps in dynamic aesthetics and a marked single- to multi-shot performance drop that widens for weaker models.

cs.CV↗

arXiv2025

Graph Embedding with Mel-spectrograms for Underwater Acoustic Target Recognition

Underwater acoustic target recognition (UATR) is extremely challenging due to the complexity of ship-radiated noise and the variability of ocean environments. Although deep learning (DL) approaches have achieved promising results, most existing models implicitly assume that underwater acoustic data lie in a Euclidean space. This assumption, however, is unsuitable for the inherently complex topology of underwater acoustic signals, which exhibit non-stationary, non-Gaussian, and nonlinear characteristics. To overcome this limitation, this paper proposes the UATR-GTransformer, a non-Euclidean DL model that integrates Transformer architectures with graph neural networks (GNNs). The model comprises three key components: a Mel patchify block, a GTransformer block, and a classification head. The Mel patchify block partitions the Mel-spectrogram into overlapping patches, while the GTransformer block employs a Transformer Encoder to capture mutual information between split patches to generate Mel-graph embeddings. Subsequently, a GNN enhances these embeddings by modeling local neighborhood relationships, and a feed-forward network (FFN) further performs feature transformation. Experiments results based on two widely used benchmark datasets demonstrate that the UATR-GTransformer achieves performance competitive with state-of-the-art methods. In addition, interpretability analysis reveals that the proposed model effectively extracts rich frequency-domain information, highlighting its potential for applications in ocean engineering.

cs.SD↗

arXiv2014

MICA: A fast short-read aligner that takes full advantage of Intel Many Integrated Core Architecture (MIC)

Background: Short-read aligners have recently gained a lot of speed by exploiting the massive parallelism of GPU. An uprising alternative to GPU is Intel MIC; supercomputers like Tianhe-2, currently top of TOP500, is built with 48,000 MIC boards to offer ~55 PFLOPS. The CPU-like architecture of MIC allows CPU-based software to be parallelized easily; however, the performance is often inferior to GPU counterparts as an MIC board contains only ~60 cores (while a GPU board typically has over a thousand cores). Results: To better utilize MIC-enabled computers for NGS data analysis, we developed a new short-read aligner MICA that is optimized in view of MICs limitation and the extra parallelism inside each MIC core. Experiments on aligning 150bp paired-end reads show that MICA using one MIC board is 4.9 times faster than the BWA-MEM (using 6-core of a top-end CPU), and slightly faster than SOAP3-dp (using a GPU). Furthermore, MICAs simplicity allows very efficient scale-up when multiple MIC boards are used in a node (3 cards give a 14.1-fold speedup over BWA-MEM). Summary: MICA can be readily used by MIC-enabled supercomputers for production purpose. We have tested MICA on Tianhe-2 with 90 WGS samples (17.47 Tera-bases), which can be aligned in an hour less than 400 nodes. MICA has impressive performance even though the current MIC is at its initial stage of development (the next generation of MIC has been announced to release in late 2014).

cs.DC↗