Bayesian Modeling of Zero-Shot Classifications for Urban Flood Detection

• URL: http://arxiv.org/abs/2503.14754v2
• Date: Wed, 26 Mar 2025 12:25:03 GMT
• Title: Bayesian Modeling of Zero-Shot Classifications for Urban Flood Detection
• Authors: Matt Franchi, Nikhil Garg, Wendy Ju, Emma Pierson,
• Abstract summary: Street scene datasets offer a promising means of detecting urban objects and incidents like street flooding.<n>A major challenge in using these datasets is their lack of reliable labels.<n>Here, we propose BayFlood, a two-stage approach which circumvents this difficulty.
• Score: 9.264147242217364
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Street scene datasets, collected from Street View or dashboard cameras, offer a promising means of detecting urban objects and incidents like street flooding. However, a major challenge in using these datasets is their lack of reliable labels: there are myriad types of incidents, many types occur rarely, and ground-truth measures of where incidents occur are lacking. Here, we propose BayFlood, a two-stage approach which circumvents this difficulty. First, we perform zero-shot classification of where incidents occur using a pretrained vision-language model (VLM). Second, we fit a spatial Bayesian model on the VLM classifications. The zero-shot approach avoids the need to annotate large training sets, and the Bayesian model provides frequent desiderata in urban settings - principled measures of uncertainty, smoothing across locations, and incorporation of external data like stormwater accumulation zones. We comprehensively validate this two-stage approach, showing that VLMs provide strong zero-shot signal for floods across multiple cities and time periods, the Bayesian model improves out-of-sample prediction relative to baseline methods, and our inferred flood risk correlates with known external predictors of risk. Having validated our approach, we show it can be used to improve urban flood detection: our analysis reveals 113,738 people who are at high risk of flooding overlooked by current methods, identifies demographic biases in existing methods, and suggests locations for new flood sensors. More broadly, our results showcase how Bayesian modeling of zero-shot LM annotations represents a promising paradigm because it avoids the need to collect large labeled datasets and leverages the power of foundation models while providing the expressiveness and uncertainty quantification of Bayesian models.

Related papers

• Confidence Aware Learning for Reliable Face Anti-spoofing [52.23271636362843]
  We propose a Confidence Aware Face Anti-spoofing model, which is aware of its capability boundary. We estimate its confidence during the prediction of each sample. Experiments show that the proposed CA-FAS can effectively recognize samples with low prediction confidence.
  arXiv  Detail & Related papers  (2024-11-02T14:29:02Z)
• Generative Edge Detection with Stable Diffusion [52.870631376660924]
  Edge detection is typically viewed as a pixel-level classification problem mainly addressed by discriminative methods. We propose a novel approach, named Generative Edge Detector (GED), by fully utilizing the potential of the pre-trained stable diffusion model. We conduct extensive experiments on multiple datasets and achieve competitive performance.
  arXiv  Detail & Related papers  (2024-10-04T01:52:23Z)
• Off to new Shores: A Dataset & Benchmark for (near-)coastal Flood Inundation Forecasting [7.4807361562214405]
  Floods are among the most common and devastating natural hazards. Recent progress in weather prediction and spaceborne flood mapping demonstrated the feasibility of anticipating extreme events. There is a critical lack of datasets and benchmarks to enable the direct forecasting of flood extent.
  arXiv  Detail & Related papers  (2024-09-27T09:51:25Z)
• Ensemble quantile-based deep learning framework for streamflow and flood prediction in Australian catchments [0.31666540219908274]
  In recent years, climate extremes such as floods have created significant environmental and economic hazards for Australia.<n>Deep learning methods have been promising for predicting extreme climate events; however, large flooding events present a critical challenge.<n>We present an ensemble quantile-based deep learning framework that addresses large-scale streamflow forecasts.
  arXiv  Detail & Related papers  (2024-07-20T23:45:04Z)
• Downstream-Pretext Domain Knowledge Traceback for Active Learning [138.02530777915362]
  We propose a downstream-pretext domain knowledge traceback (DOKT) method that traces the data interactions of downstream knowledge and pre-training guidance. DOKT consists of a traceback diversity indicator and a domain-based uncertainty estimator. Experiments conducted on ten datasets show that our model outperforms other state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-07-20T01:34:13Z)
• Lazy Layers to Make Fine-Tuned Diffusion Models More Traceable [70.77600345240867]
  A novel arbitrary-in-arbitrary-out (AIAO) strategy makes watermarks resilient to fine-tuning-based removal. Unlike the existing methods of designing a backdoor for the input/output space of diffusion models, in our method, we propose to embed the backdoor into the feature space of sampled subpaths. Our empirical studies on the MS-COCO, AFHQ, LSUN, CUB-200, and DreamBooth datasets confirm the robustness of AIAO.
  arXiv  Detail & Related papers  (2024-05-01T12:03:39Z)
• Rapid Flood Inundation Forecast Using Fourier Neural Operator [77.30160833875513]
  Flood inundation forecast provides critical information for emergency planning before and during flood events. High-resolution hydrodynamic modeling has become more accessible in recent years, however, predicting flood extents at the street and building levels in real-time is still computationally demanding. We present a hybrid process-based and data-driven machine learning (ML) approach for flood extent and inundation depth prediction.
  arXiv  Detail & Related papers  (2023-07-29T22:49:50Z)
• Toward Data-Driven Glare Classification and Prediction for Marine Megafauna Survey [0.3694429692322631]
  Critically endangered species in Canadian North Atlantic waters are systematically surveyed to estimate species populations. Due to its impact on policy, population accuracy is important. This paper lays the foundation towards a data-driven glare modelling system, which will allow surveyors to preemptively minimize glare.
  arXiv  Detail & Related papers  (2023-03-03T18:46:19Z)
• An evaluation of deep learning models for predicting water depth evolution in urban floods [59.31940764426359]
  We compare different deep learning models for prediction of water depth at high spatial resolution. Deep learning models are trained to reproduce the data simulated by the CADDIES cellular-automata flood model. Our results show that the deep learning models present in general lower errors compared to the other methods.
  arXiv  Detail & Related papers  (2023-02-20T16:08:54Z)
• Determination of building flood risk maps from LiDAR mobile mapping data [11.128919681798664]
  Flood simulations can provide early warnings for areas and buildings at risk of flooding. Knowing the heights of facade openings helps to identify places that are more susceptible to water ingress. This research presents a new process for the extraction of windows and doors from LiDAR mobile mapping data.
  arXiv  Detail & Related papers  (2022-01-14T15:36:08Z)
• Contributions to Large Scale Bayesian Inference and Adversarial Machine Learning [0.0]
  The rampant adoption of ML methodologies has revealed that models are usually adopted to make decisions without taking into account the uncertainties in their predictions. We believe that developing ML systems that take into predictive account uncertainties and are robust against adversarial examples is a must for real-world tasks.
  arXiv  Detail & Related papers  (2021-09-25T23:02:47Z)

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.