video surveillance operators

Abstract: Many questions about a video concern one physical object. Where does the ball end up after the cups are shuffled? Which card is the queen? How often was the mug covered? Video-language models (Video-LMs) recognise such objects in single frames but do not keep them apart over time. On real shell-game videos, open models from 8B to 235B parameters, their thinking variants, a reinforcement-learned tool agent and three training-free tool agents all stay within ten points of chance. We separate semantic identity binding from temporal identity maintenance. An object's identity should be fixed at a moment when the evidence for it is visible, and carried afterwards by a tracker. \emph{Track, Then Bind} implements this principle without training. A promptable video tracker follows every candidate object and draws it with a track number. A frozen vision-language model answers a static question about the identifying evidence, which binds the target to a track. The answer is then derived from the tracked state in text, without asking the model to reinterpret the original video during read-out. A single configuration covers all benchmarks. With Qwen3-VL-32B it solves both VET-Bench tasks, where a fine-tuned trajectory model reaches 92\% and 90\% and the same VLM voting over 16 samples stays at 30--32\%. It answers 55.4\% of the hardest Perception Test cup games, where the best published model answers 41.5\%, and, using track-derived timelines, 48.6\% of TOC-Bench questions, against 37.1\% for the same model answering directly. Drawing the same tracks on the video and asking the question directly falls 12--19 points short of the agent on real cup games.

Abstract: Multimodal large language models (MLLMs) have made rapid progress in general visual understanding, yet they still struggle with fine-grained instance-level consistency verification, where the goal is to determine whether two images depict the exact same physical object under varying viewpoints, illumination, and backgrounds. This task requires both global semantic perception and precise examination of subtle local evidence, since semantically similar objects may differ only in small suspicious regions. To address this challenge, we propose LocalAgent, a collaborative agentic framework that integrates the complementary strengths of a generalist MLLM and lightweight specialist verifiers. The MLLM first reasons globally to identify suspicious local regions, which are then delegated to a Local Evidence Verifier (LEV) toolkit for fine-grained appearance and geometric matching. LEV comprises Local Appearance Consistency (LAC), which estimates appearance similarity under illumination and color changes, and Local Geometric Correspondence (LGC), which estimates spatial correspondence under viewpoint changes. LEV returns structured scalar evidence that the MLLM integrates for the final decision. We further introduce an auxiliary-capped reinforcement fine-tuning (RFT) strategy, in which answer correctness dominates auxiliary tool-use and format signals, enabling the policy to learn when verification is necessary. Experiments show that LocalAgent improves mean accuracy by 4.9% over the strongest CoT-RFT baseline while invoking a verifier on only 12.13% of samples and preserving the general multimodal capabilities of the underlying MLLM. The open-source repository is available at https://github.com/Firefly-zoo/LocalAgent-main.

Abstract: Online scene change detection requires integrating incoming observations to determine which regions have changed relative to a reference scene. We update change beliefs as new images arrive using generalised Bayesian evidence fusion. However, the prior probability of change and evidence models are unknown beforehand, and numerical Bayesian inference can be costly. We address these challenges by learning offline from automatic edits to a reference 3D Gaussian Splatting (3DGS) model. These edits generate labelled training views without manual annotations. Together with unchanged reference observations, they provide supervision for learning scene-specific evidence models and estimating the effective pixel prior and response statistics. Online prediction combines current evidence with historical support, while closed-form local Bayes factors, parameterized by the learned response statistics, update an accumulated 3D Gaussian field. Object-guided refinement produces the final change masks. Our method achieves state-of-the-art accuracy on PASLCD and CCS3D, with $67.64\%$ and $75.82\%$ mIoU, respectively. On PASLCD, the complete pipeline runs at $17.2$\,FPS over complete query sequences, the fastest among the evaluated online methods.

Abstract: Person re-identification (ReID) ranks an image gallery using either an image query or a textual query. In image-to-image (I2I) ReID, viewpoint, illumination, occlusion, and background changes can make distractors appear more similar than the true match. In text-to-image (T2I) ReID, incomplete or ambiguous descriptions may omit key visual evidence. We study this problem for vision-language model (VLM) based ReID, where attributes such as clothing color and carried objects provide useful fine-grained matching cues. However, directly using attribute-prompt similarities as retrieval scores is unreliable because templated prompts can be biased, correlated, and poorly calibrated. We therefore propose \textbf{PR}ompt-guided \textbf{A}ttribute-based \textbf{I}nference-time \textbf{S}imilarity \textbf{E}nhancement (\texttt{PRAISE}), a query-side test-time adaptation framework for VLM-based ReID. Instead of treating prompts as scores, \texttt{PRAISE} uses a transferable attribute-prompt bank to construct an \emph{Attribute-Conditioned Subspace} (ACS): a query-dependent low-rank subspace induced by the most compatible attribute prompts. ACS provides a controlled attribute-conditioned geometry for reshaping query--gallery affinities and guiding lightweight inference-time updates without identity or attribute labels. Experiments across multiple I2I- and T2I-ReID benchmarks show that \texttt{PRAISE} consistently improves VLM retrieval backbones and outperforms both generic and ReID-specific TTA baselines.

Abstract: Person re-identification is typically addressed by learning identity-discriminative embeddings from individual person crops. However, player re-identification in broadcast sports videos presents a unique challenge: players from the same team often share similar appearances, while video sequences provide repeated observations of the same players under different viewpoints and surrounding contexts. Existing crop-based approaches largely ignore this contextual information, which can be critical for resolving visually similar identities. Here, we propose ContextReID, a contextualized embedding framework that represents a player by jointly considering other players in the same frame and observations from different frames of the broadcast sequence. By incorporating both co-occurring players and cross-frame observations, the model learns identity features relative to the surrounding visual context rather than relying only on isolated appearance cues. We implement this framework with a vision-language model that takes sets of player crops as input and generates context-aware embeddings for individual players. A combined contextual learning and retrieval objective encourages the model to capture cross-frame identity relationships while maintaining discriminative embeddings for player matching. Experiments across three multi-object tracking benchmarks show that ContextReID improves player retrieval over conventional single-crop embedding approaches in both Rank-1 and mAP. These results highlight the utility of spatial and temporal context for reliable player identification in broadcast sports videos.

Abstract: Intermediate-layer features from multimodal large language models have shown strong potential for video anomaly detection (VAD), yet the origin of their discriminative power remains unclear. We study this question using sparse mixture-of-experts (MoE) models, whose explicit expert structure and sparse activation make their internal computation easier to inspect. With a fully frozen backbone and no additional training, we find that anomaly-related evidence is concentrated in a small set of experts. These experts recur across layers, spontaneously specialize in different anomaly types, and together form a dynamic routing subnetwork. We further show that the output channels most strongly influenced by these experts are also the hidden dimensions that contain the most anomaly-relevant information. Routing statistics can therefore serve as an internal anomaly cue that complements semantic features.Based on these findings, we propose RoMod, an efficient VAD framework trained with only \(5\%\) of weakly labeled videos. RoMod includes a Routing-Modulated Fusion module, RoMF, and a Routing-aware Temporal Network, RoTN. RoMF uses routing signals to adaptively recalibrate hidden semantic channels. Its design also prevents the routing branch from bypassing semantic features and making predictions on its own. RoTN captures the temporal evolution of anomalies from onset to persistence and termination. Experiments on three benchmarks show that RoMod achieves state-of-the-art performance while running substantially faster than dense backbones of comparable size.

Abstract: We describe our winning entry to the Waymo Open Dataset 2D Video Panoptic Segmentation Challenge. The task asks for a semantic class at every pixel of every frame and, for countable objects, an identity that holds across 100 frames and across five overlapping cameras. We build on DVIS++, a cascade of a segmenter, a tracker, and a refiner, as our baseline. We propose hard region supervision (HRS) to improve the baseline. In particular, we use the baseline to define the hard region as where it makes mistakes, and design a loss and an auxiliary prediction head for this region. The auxiliary head is used only in training and removed at test time, so at inference the model trained with HRS has the same architecture as the baseline. In addition, we propose three test-time steps that further improve the results: a two-model ensemble, a merge of the segmenter's output into the final panoptic map, and cross-camera identity linking. On the challenge test set, our entry reaches 0.3547 wSTQ, 0.2071 wAQ, and 0.6075 mIoU, ranking first on all three metrics. It is 3.6 wSTQ points ahead of the second entry and 2.4 points ahead of our DVIS++ baseline.

Abstract: The ByteTrack algorithm is a widely used and computationally efficient multi-object tracking architecture. Its core innovation lies in the combination of lenient bounding box associations with tracklet similarity matching to robustly deal with object occlusions. However, this strategy is nevertheless vulnerable to erroneous track reclassification and identity switching, as detection confidence scores dictate association priority. To address this, I present a simple enhancement of the ByteTrack architecture, named ByteTraX, that optimises track continuity via a single unified matching threshold, while penalising identity switches through stringent track initiation criteria. This approach achieves consistently improved performance across a range of diverse benchmarks including GMOT-40, LC-MOT, SportsMOT, TeamTrack, DAMUNT, and DeepSea-MOT, while simultaneously increasing processing speed by >10%. Specifically, results demonstrate a >40% reduction in identity switches, accompanied by mean increases in HOTA of 3.6, IDF1 of 5.6, and FPS of 6.3. As such, adoption of the ByteTraX algorithm has the potential to substantially enhance tracking performance over the ByteTrack baseline, while retaining the efficiency needed for real-time deployment. To facilitate usage, I provide the source code, integration functionality for the YOLO family of object detection models, and deployment instructions via an open source repository.

Abstract: Online multi camera 3D tracking must maintain scene global identities across synchronized views, yet query-based trackers carry these identities only implicitly in the instance bank, where they fragment upon query interruption. We present an online architecture that recovers association accuracy by predicting IDs explicitly over recurrent sparse queries. An outside-in Sparse4D detector fuses calibrated views into world frame 3D detections while propagating a sparse query bank, and a causal MOTIP ID decoder associates detections against a finite trajectory memory. We adapt MOTIP's relative-ID prediction and recycled slot runtime to globally fused 3D observations, and introduce metric spatial gating and proximity based newborn recovery. On the official 2026 AI City Challenge Track 1 test set, our method raises HOTA from 29.63 with native instance bank identities to 38.01, primarily through an AssA increase from 20.83 to 31.10, and ranks third on the public leaderboard. Full-sequence validation over all 9,000 frames of each scene shows that decoupled ID training improves HOTA over native identities, whereas continuing detector training alongside the detached ID objective produces scene-dependent gains and losses.