XD-Violence
PulseAugur coverage of XD-Violence — every cluster mentioning XD-Violence across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New audit reveals limited gains from frame selection in long-video QA
A new audit methodology called VideoRouteAudit has been developed to dissect the performance gains in long-video question answering tasks. The audit reveals that traditional frame selection methods offer minimal improve…
-
New SAAF framework achieves efficient zero-shot video anomaly detection
Researchers have developed a new framework called Signal-Aware Adaptive Fusion (SAAF) for efficient zero-shot video anomaly detection. SAAF addresses the computational limitations of current Vision-Language Models by em…
-
MoG-VLM: Motion-Guided Vision-Language Models Enhance Video Anomaly Detection
Researchers have developed MoG-VLM, a novel framework for efficient video anomaly detection that utilizes motion to guide the inference process of vision-language models. This approach selectively activates VLM analysis…
-
New AI framework enhances multi-label video safety detection
Researchers have developed a new framework called Adaptive Tversky Policy Optimization (ATPO) to improve multi-label video safety detection. This approach uses a reinforcement learning framework with an Adaptive Tversky…
-
AI model predicts violence using pre-incident behavioral cues
Researchers have developed a multimodal learning approach to anticipate violence by analyzing pre-incident behavioral cues. The study utilizes the XD-Violence dataset, focusing on short-horizon risk recognition from vid…
-
Research questions violence detection benchmarks, finds artifact influence
A new research paper questions the effectiveness of interaction representations in weakly-supervised violence detection, finding that coarse geometric representations perform as well as or better than more complex pose-…
-
New research tackles video anomaly detection with causal models and improved evaluation
Researchers are exploring new methods for video anomaly detection, focusing on improving efficiency and accuracy. One paper introduces a strictly causal streaming anomaly detector using a Mamba-style state-space model t…
-
New framework enhances video anomaly detection with spatial localization
Researchers have developed SST-WSVADL, a new framework designed to improve weakly supervised video anomaly detection (WSVAD) by focusing on spatial localization rather than just temporal cues. This approach aims to miti…
-
New research addresses definition blindness in video anomaly detection
Two new research papers tackle the challenge of open-world video anomaly detection, where systems must identify user-defined abnormal events. The first paper, "Rethinking Open-World Video Anomaly Detection," introduces …
-
New PULS system anticipates video anomalies using semantic world-model pipeline
Researchers have developed PULS (Predictive Unified Latent Space), a novel pipeline for continuous video anomaly detection that moves beyond reactive methods. PULS consists of a KSD Bridge, which translates physical ten…
-
MemoVAD enables efficient edge video anomaly detection with VLM
Researchers have developed MemoVAD, a novel framework for resource-efficient video anomaly detection on edge devices. This system uses a combination of edge and cloud processing, with a unique uncertainty-aware gating p…
-
New dataset ExtrAnom enhances women's safety video anomaly detection
Researchers have introduced the ExtrAnom dataset, a new multi-modal benchmark designed to improve video anomaly detection (VAD) specifically for women's safety. The dataset contains 1001 videos, with 501 labeled as anom…
-
New framework uses frozen VLM for training-free video anomaly detection
Researchers have developed CoReVAD, a novel framework for detecting anomalies in videos without requiring task-specific training. This approach leverages a single, frozen Vision-Language Model (VLM) to generate both ano…
-
New architecture boosts audio language models' attention to salient sounds
Researchers have developed NAACA, a novel architecture designed to improve how audio language models process long audio recordings. NAACA uses a training-free approach with an Oscillatory Working Memory (OWM) to filter …
-
LLMs enhance video anomaly detection with reasoning and spatial grounding
Researchers have developed VANGUARD, a novel framework that integrates video anomaly detection with multimodal large language models. This system not only identifies anomalies but also provides interpretable chain-of-th…