Two-Pass Zero-Shot Temporal-Spatial Grounding of Rare Traffic Events in Surveillance Video

• URL: http://arxiv.org/abs/2605.01512v1
• Date: Sat, 02 May 2026 16:05:44 GMT
• Title: Two-Pass Zero-Shot Temporal-Spatial Grounding of Rare Traffic Events in Surveillance Video
• Abstract summary: Grounding traffic accidents in real CCTV footage is a rare-event problem where training on labeled accident video is often prohibited.<n>We present a no-fine-tuning pipeline that elicits this joint output from frozen vision-language models.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Grounding traffic accidents in real CCTV footage is a rare-event problem where training on labeled accident video is often prohibited, yet accurate joint localization in time, space, and collision type is required. We present a no-fine-tuning pipeline that elicits this joint output from frozen vision-language models through two ideas. First, a coarse-to-fine two-pass decomposition: a full-video pass at 1 fps produces a coarse (t, x, y, c) tuple, then a second pass at 5 fps within a +/- 3 s window refines time and location, with two deterministic confidence gates that revert to the coarse estimate on boundary hedges or edge-clamped coordinates. Second, a specialist role assignment: Qwen3-VL-Plus handles grounding, Gemini 3.1 Flash-Lite handles typing on a centered video clip. On the ACCIDENT@CVPR 2026 benchmark (2,027 real CCTV videos) we reach ACC^S = 0.539 (95% CI [0.525, 0.553]): +0.127 over the benchmark paper's best-of-baselines oracle (0.412), +0.143 over the strongest single-VLM baseline (Molmo-7B, 0.396), and +0.250 over the naive baseline (0.289). The VLM path uses up to three API calls per video (17% fall back to physics on API failures); the full run costs ~$20.

Related papers

• VC-Attention: Value Smoothing and Softmax Casting for Low-bit Attention [47.30761996669105]
  We propose VC-Attention, a training-free low-bit attention framework.<n>It addresses both by smoothing value tokens by a probability Cast. V-Smooth reorders value tokens by lightweight clustering online, so the tokens in a fused hardware block quantize well together.<n>It maps log-domain scores directly to E4M3 probability codes with one fused multiply-add, eliminating the FP32 exponential and the format conversion.
  arXiv  Detail & Related papers  (2026-09-14T16:17:35Z)
• Strictly Causal Streaming Video Anomaly Detection with a Theoretically-Grounded State-Space Core [2.648500779572419]
  We introduce a causal streaming anomaly detector with no lookahead and no clip buffering.<n>Its temporal core is a diagonal linear state space recurrence with an input and state dependent decay gate.<n>We derive a closed form relationship between the recurrence decay spectrum and both detection delay and the shortest anomaly it can reliably capture.
  arXiv  Detail & Related papers  (2026-08-25T16:52:32Z)
• Zero-Shot Traffic Accident Detection via a Coarse-to-Fine VLM-Tracking Pipeline [0.10874100424278173]
  ACCIDENT @ CVPR benchmark evaluates exactly this joint prediction under a strict constraint.<n>We introduce a training-free, two-pass coarse-to-fine pipeline that pairs a frozen Qwen3-VL-32B-Instruct vision-language model with YOLO11x object detection and BoT-SORT tracking.<n>On the official 2,027-clip real-CCTV test set, our system achieves a three-way harmonic mean score of 0.504.
  arXiv  Detail & Related papers  (2026-08-09T19:08:56Z)
• The Low Frequency Trap: Video Language Models Fail at Simple Event Bookkeeping [70.07961857926377]
  We introduce trace-grounded parametric profiling for event counting in three controlled video tasks.<n>Across 2,190 videos, we vary event count N and frequency F while holding rendering fixed.<n>At an 80% reliability threshold, Gemini 3.6 Flash reliably counts persistent state transitions up to 12 events at 0.5 and 1.0 Hz, yet demonstrates no reliable positive-count region for transient blinking events.
  arXiv  Detail & Related papers  (2026-08-06T17:57:06Z)
• A global predicted-fMRI drive signal from TRIBE does not predict YouTube replay heatmaps [0.0]
  multimodal brain-encoding models predict fMRI responses to naturalistic video with high accuracy.<n>We run TRIBE on 48 YouTube videos and reduce its predicted cortical response to a per-second engagement curve.<n>Running the probe on TRIBE's input streams reveals at most a small, borderline visual-stream signal.<n>We release code, a video-ID manifest, and a heatmap-acquisition method robust to YouTube's SABR streaming.
  arXiv  Detail & Related papers  (2026-07-01T19:03:17Z)
• VisualClaw: A Real-Time, Personalized Agent for the Physical World [76.89211120690028]
  We present VisualClaw, a self-evolving multimodal agent built around two principles.<n>First, hybrid encoding reduces deployment cost by filtering less informative streaming frames with a cascaded gate.<n>Second, skill evolution lets the agent learn from failures, producing skill-bank updates that help future questions.
  arXiv  Detail & Related papers  (2026-06-15T06:58:22Z)
• Scaling Parallel Sequence Models to Foundation-Scale Vision Encoders [123.58723804218151]
  Vision foundation models are bottlenecked by the quadratic cost of self-attention.<n>We present C-GSPN based on 2D spatial propagation propagation.
  arXiv  Detail & Related papers  (2026-05-30T14:29:43Z)
• Multi-Stage VLM Pipeline for Zero-Shot Traffic Accident Understanding [0.0]
  We present the 1st-place solution to the ACCIDENT challenge at the CVPR 2026 AUTOPILOT Workshop.<n>The challenge asks for zero-shot prediction of accident timing, impact centroid, and collision type from CCTV footage.<n>We build a three-stage pipeline (full-video joint prediction, time refinement, and single-frame grounding of the impact centroid), run the same pipeline a second time on a 235B Mixture-of-Experts sibling, blend the two outputs 9:1, and finally snap each predicted point onto the nearest vehicle detection.
  arXiv  Detail & Related papers  (2026-05-28T04:00:26Z)
• Causal Forcing++: Scalable Few-Step Autoregressive Diffusion Distillation for Real-Time Interactive Video Generation [50.55853275866995]
  Real-time interactive video generation requires low-latency, streaming, and controllable rollout.<n>In this paper, we study a more aggressive setting: frame-wise autoregression with only 1--2 sampling steps.<n>We propose textbfCausal Forcing++, a principled and scalable pipeline.
  arXiv  Detail & Related papers  (2026-05-14T17:46:36Z)
• Instance-level Visual Active Tracking with Occlusion-Aware Planning [61.982298426203165]
  Visual Active Tracking (VAT) aims to control cameras to follow a target in 3D space.<n>VAT faces two key bottlenecks in real-world deployment: confusion from visually similar distractors and severe failure under occlusions.<n>We propose OA-VAT, a unified pipeline with three complementary modules.
  arXiv  Detail & Related papers  (2026-04-23T09:11:50Z)
• A Modular Zero-Shot Pipeline for Accident Detection, Localization, and Classification in Traffic Surveillance Video [0.0]
  We describe a zero-shot pipeline developed for the ACCIDENT @ CVPR 2026 challenge.<n>The challenge requires predicting when, where, and what type of traffic accident occurs in surveillance video, without labeled real-world training data.<n>Our method separates the problem into three independent modules.
  arXiv  Detail & Related papers  (2026-04-05T05:18:19Z)
• Scaling DoRA: High-Rank Adaptation via Factored Norms and Fused Kernels [83.99688944263843]
  Weight-De Low-Rank Adaptation (DoRA) extends LoRA by decoupling weight magnitude from direction.<n>At d_in = 8192 and rank r = 384, a single module's norm requires about 512 MB of transient working memory in bf16.<n>A factored norm decomposes the squared norm into base, cross, and Gram terms computable through O(d_out r + r2) intermediates, eliminating the dense product.
  arXiv  Detail & Related papers  (2026-03-23T17:57:24Z)
• ParkingTwin: Training-Free Streaming 3D Reconstruction for Parking-Lot Digital Twins [8.681741531117366]
  ParkingTwin is a training-free, lightweight system for online streaming 3D reconstruction.<n>It runs at 30+ FPS on an entry-level GTX 1660.
  arXiv  Detail & Related papers  (2026-01-20T08:03:58Z)
• Semantic-Drive: Democratizing Long-Tail Data Curation via Open-Vocabulary Grounding and Neuro-Symbolic VLM Consensus [0.0]
  We introduce Semantic-Drive, a local-first, neuro-symbolic framework for semantic data mining.<n>Our approach decouples perception into two stages: (1) Grounding via a real-time open-vocabulary detector (YOLOE) to anchor attention, and (2) Cognitive Analysis via a Reasoning VLM that performs forensic scene analysis.
  arXiv  Detail & Related papers  (2025-12-12T20:07:04Z)
• POVQA: Preference-Optimized Video Question Answering with Rationales for Data Efficiency [3.4998703934432682]
  Recent advancements in large context/long video question answering have allowed VQA tasks to have context window of 1500+ frames.<n>We introduce POVQA, a data-efficient pipeline that compresses each second of video into a single temporally pooled image.
  arXiv  Detail & Related papers  (2025-10-01T15:15:36Z)
• From Play to Replay: Composed Video Retrieval for Temporally Fine-Grained Videos [48.666667545084835]
  Composed Video Retrieval (CoVR) retrieves a target video given a query video and a modification text describing the intended change.<n>We introduce TF-CoVR, the first large-scale benchmark dedicated to temporally fine-grained CoVR.<n> TF-CoVR focuses on gymnastics and diving and provides 180K triplets drawn from FineGym and FineDiving.
  arXiv  Detail & Related papers  (2025-06-05T17:31:17Z)
• VideoFlow: Exploiting Temporal Cues for Multi-frame Optical Flow Estimation [61.660040308290796]
  VideoFlow is a novel optical flow estimation framework for videos. We first propose a TRi-frame Optical Flow (TROF) module that estimates bi-directional optical flows for the center frame in a three-frame manner. With the iterative flow estimation refinement, the information fused in individual TROFs can be propagated into the whole sequence via MOP.
  arXiv  Detail & Related papers  (2023-03-15T03:14:30Z)
• DualFormer: Local-Global Stratified Transformer for Efficient Video Recognition [140.66371549815034]
  We propose a new transformer architecture, termed DualFormer, which can effectively and efficiently perform space-time attention for video recognition. We show that DualFormer sets new state-of-the-art 82.9%/85.2% top-1 accuracy on Kinetics-400/600 with around 1000G inference FLOPs which is at least 3.2 times fewer than existing methods with similar performances.
  arXiv  Detail & Related papers  (2021-12-09T03:05:19Z)
• Spatiotemporal Contrastive Video Representation Learning [87.56145031149869]
  We present a self-supervised Contrastive Video Representation Learning (CVRL) method to learn visual representations from unlabeled videos. Our representations are learned using a contrasttemporalive loss, where two augmented clips from the same short video are pulled together in the embedding space. We study what makes for good data augmentations for video self-supervised learning and find that both spatial and temporal information are crucial.
  arXiv  Detail & Related papers  (2020-08-09T19:58:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.