HourVideo: 1-Hour Video-Language Understanding

We present HourVideo, a benchmark dataset for hour-long video-language understanding. Our dataset consists of a novel task suite comprising summarization, perception (recall, tracking), visual reasoning (spatial, temporal, predictive, causal, counterfactual), and navigation (room-to-room, object retrieval) tasks. HourVideo includes 500 manually curated egocentric videos from the Ego4D dataset, spanning durations of 20 to 120 minutes, and features 12,976 high-quality, five-way multiple-choice questions. Benchmarking results reveal that multimodal models, including GPT-4 and LLaVA-NeXT, achieve marginal improvements over random chance. In stark contrast, human experts significantly outperform the state-of-the-art long-context multimodal model, Gemini Pro 1.5 (85.0% vs. 37.3%), highlighting a substantial gap in multimodal capabilities. Our benchmark, evaluation toolkit, prompts, and documentation are available at https://hourvideo.stanford.edu

Eyes on the Road: State-of-the-Art Video Question Answering Models Assessment for Traffic Monitoring Tasks

Recent advances in video question answering (VideoQA) offer promising applications, especially in traffic monitoring, where efficient video interpretation is critical. Within ITS, answering complex, real-time queries like "How many red cars passed in the last 10 minutes?" or "Was there an incident between 3:00 PM and 3:05 PM?" enhances situational awareness and decision-making. Despite progress in vision-language models, VideoQA remains challenging, especially in dynamic environments involving multiple objects and intricate spatiotemporal relationships. This study evaluates state-of-the-art VideoQA models using non-benchmark synthetic and real-world traffic sequences. The framework leverages GPT-4o to assess accuracy, relevance, and consistency across basic detection, temporal reasoning, and decomposition queries. VideoLLaMA-2 excelled with 57% accuracy, particularly in compositional reasoning and consistent answers. However, all models, including VideoLLaMA-2, faced limitations in multi-object tracking, temporal coherence, and complex scene interpretation, highlighting gaps in current architectures. These findings underscore VideoQA's potential in traffic monitoring but also emphasize the need for improvements in multi-object tracking, temporal reasoning, and compositional capabilities. Enhancing these areas could make VideoQA indispensable for incident detection, traffic flow management, and responsive urban planning. The study's code and framework are open-sourced for further exploration: https://github.com/joe-rabbit/VideoQA_Pilot_Study

Hatching-Box: Monitoring the Rearing Process of Drosophila Using an Embedded Imaging and in-vial Detection System

In this paper we propose the Hatching-Box, a novel imaging and analysis system to automatically monitor and quantify the developmental behavior of Drosophila in standard rearing vials and during regular rearing routines, rendering explicit experiments obsolete. This is achieved by combining custom tailored imaging hardware with dedicated detection and tracking algorithms, enabling the quantification of larvae, filled/empty pupae and flies over multiple days. Given the affordable and reproducible design of the Hatching-Box in combination with our generic client/server-based software, the system can easily be scaled to monitor an arbitrary amount of rearing vials simultaneously. We evaluated our system on a curated image dataset comprising nearly 470,000 annotated objects and performed several studies on real world experiments. We successfully reproduced results from well-established circadian experiments by comparing the eclosion periods of wild type flies to the clock mutants $\textit{per}^{short}$, $\textit{per}^{long}$ and $\textit{per}^0$ without involvement of any manual labor. Furthermore we show, that the Hatching-Box is able to extract additional information about group behavior as well as to reconstruct the whole life-cycle of the individual specimens. These results not only demonstrate the applicability of our system for long-term experiments but also indicate its benefits for automated monitoring in the general cultivation process.

DROID-Splat: Combining end-to-end SLAM with 3D Gaussian Splatting

Recent progress in scene synthesis makes standalone SLAM systems purely based on optimizing hyperprimitives with a Rendering objective possible \cite{monogs}. However, the tracking performance still lacks behind traditional \cite{orbslam} and end-to-end SLAM systems \cite{droid}. An optimal trade-off between robustness, speed and accuracy has not yet been reached, especially for monocular video. In this paper, we introduce a SLAM system based on an end-to-end Tracker and extend it with a Renderer based on recent 3D Gaussian Splatting techniques. Our framework \textbf{DroidSplat} achieves both SotA tracking and rendering results on common SLAM benchmarks. We implemented multiple building blocks of modern SLAM systems to run in parallel, allowing for fast inference on common consumer GPU's. Recent progress in monocular depth prediction and camera calibration allows our system to achieve strong results even on in-the-wild data without known camera intrinsics. Code will be available at \url{https://github.com/ChenHoy/DROID-Splat}.

First Place Solution to the ECCV 2024 ROAD++ Challenge @ ROAD++ Atomic Activity Recognition 2024

This report presents our team's technical solution for participating in Track 3 of the 2024 ECCV ROAD++ Challenge. The task of Track 3 is atomic activity recognition, which aims to identify 64 types of atomic activities in road scenes based on video content. Our approach primarily addresses the challenges of small objects, discriminating between single object and a group of objects, as well as model overfitting in this task. Firstly, we construct a multi-branch activity recognition framework that not only separates different object categories but also the tasks of single object and object group recognition, thereby enhancing recognition accuracy. Subsequently, we develop various model ensembling strategies, including integrations of multiple frame sampling sequences, different frame sampling sequence lengths, multiple training epochs, and different backbone networks. Furthermore, we propose an atomic activity recognition data augmentation method, which greatly expands the sample space by flipping video frames and road topology, effectively mitigating model overfitting. Our methods rank first in the test set of Track 3 for the ROAD++ Challenge 2024, and achieve 69% mAP.

FG-PE: Factor-graph Approach for Multi-robot Pursuit-Evasion

With the increasing use of robots in daily life, there is a growing need to provide robust collaboration protocols for robots to tackle more complicated and dynamic problems effectively. This paper presents a novel, factor graph-based approach to address the pursuit-evasion problem, enabling accurate estimation, planning, and tracking of an evader by multiple pursuers working together. It is assumed that there are multiple pursuers and only one evader in this scenario. The proposed method significantly improves the accuracy of evader estimation and tracking, allowing pursuers to capture the evader in the shortest possible time and distance compared to existing techniques. In addition to these primary objectives, the proposed approach effectively minimizes uncertainty while remaining robust, even when communication issues lead to some messages being dropped or lost. Through a series of comprehensive experiments, this paper demonstrates that the proposed algorithm consistently outperforms traditional pursuit-evasion methods across several key performance metrics, such as the time required to capture the evader and the average distance traveled by the pursuers. Additionally, the proposed method is tested in real-world hardware experiments, further validating its effectiveness and applicability.

FastTrackTr:Towards Fast Multi-Object Tracking with Transformers

Transformer-based multi-object tracking (MOT) methods have captured the attention of many researchers in recent years. However, these models often suffer from slow inference speeds due to their structure or other issues. To address this problem, we revisited the Joint Detection and Tracking (JDT) method by looking back at past approaches. By integrating the original JDT approach with some advanced theories, this paper employs an efficient method of information transfer between frames on the DETR, constructing a fast and novel JDT-type MOT framework: FastTrackTr. Thanks to the superiority of this information transfer method, our approach not only reduces the number of queries required during tracking but also avoids the excessive introduction of network structures, ensuring model simplicity. Experimental results indicate that our method has the potential to achieve real-time tracking and exhibits competitive tracking accuracy across multiple datasets.

DINTR: Tracking via Diffusion-based Interpolation

Object tracking is a fundamental task in computer vision, requiring the localization of objects of interest across video frames. Diffusion models have shown remarkable capabilities in visual generation, making them well-suited for addressing several requirements of the tracking problem. This work proposes a novel diffusion-based methodology to formulate the tracking task. Firstly, their conditional process allows for injecting indications of the target object into the generation process. Secondly, diffusion mechanics can be developed to inherently model temporal correspondences, enabling the reconstruction of actual frames in video. However, existing diffusion models rely on extensive and unnecessary mapping to a Gaussian noise domain, which can be replaced by a more efficient and stable interpolation process. Our proposed interpolation mechanism draws inspiration from classic image-processing techniques, offering a more interpretable, stable, and faster approach tailored specifically for the object tracking task. By leveraging the strengths of diffusion models while circumventing their limitations, our Diffusion-based INterpolation TrackeR (DINTR) presents a promising new paradigm and achieves a superior multiplicity on seven benchmarks across five indicator representations.

V-CAS: A Realtime Vehicle Anti Collision System Using Vision Transformer on Multi-Camera Streams

This paper introduces a real-time Vehicle Collision Avoidance System (V-CAS) designed to enhance vehicle safety through adaptive braking based on environmental perception. V-CAS leverages the advanced vision-based transformer model RT-DETR, DeepSORT tracking, speed estimation, brake light detection, and an adaptive braking mechanism. It computes a composite collision risk score based on vehicles' relative accelerations, distances, and detected braking actions, using brake light signals and trajectory data from multiple camera streams to improve scene perception. Implemented on the Jetson Orin Nano, V-CAS enables real-time collision risk assessment and proactive mitigation through adaptive braking. A comprehensive training process was conducted on various datasets for comparative analysis, followed by fine-tuning the selected object detection model using transfer learning. The system's effectiveness was rigorously evaluated on the Car Crash Dataset (CCD) from YouTube and through real-time experiments, achieving over 98% accuracy with an average proactive alert time of 1.13 seconds. Results indicate significant improvements in object detection and tracking, enhancing collision avoidance compared to traditional single-camera methods. This research demonstrates the potential of low-cost, multi-camera embedded vision transformer systems to advance automotive safety through enhanced environmental perception and proactive collision avoidance mechanisms.

CamI2V: Camera-Controlled Image-to-Video Diffusion Model

Recently, camera pose, as a user-friendly and physics-related condition, has been introduced into text-to-video diffusion model for camera control. However, existing methods simply inject camera conditions through a side input. These approaches neglect the inherent physical knowledge of camera pose, resulting in imprecise camera control, inconsistencies, and also poor interpretability. In this paper, we emphasize the necessity of integrating explicit physical constraints into model design. Epipolar attention is proposed for modeling all cross-frame relationships from a novel perspective of noised condition. This ensures that features are aggregated from corresponding epipolar lines in all noised frames, overcoming the limitations of current attention mechanisms in tracking displaced features across frames, especially when features move significantly with the camera and become obscured by noise. Additionally, we introduce register tokens to handle cases without intersections between frames, commonly caused by rapid camera movements, dynamic objects, or occlusions. To support image-to-video, we propose the multiple guidance scale to allow for precise control for image, text, and camera, respectively. Furthermore, we establish a more robust and reproducible evaluation pipeline to solve the inaccuracy and instability of existing camera control measurement. We achieve a 25.5\% improvement in camera controllability on RealEstate10K while maintaining strong generalization to out-of-domain images. Only 24GB and 12GB are required for training and inference, respectively. We plan to release checkpoints, along with training and evaluation codes. Dynamic videos are best viewed at \url{https://zgctroy.github.io/CamI2V}.