Visual Haystacks: Answering Harder Questions About Sets of Images

Recent advancements in Large Multimodal Models (LMMs) have made significant progress in the field of single-image visual question answering. However, these models face substantial challenges when tasked with queries that span extensive collections of images, similar to real-world scenarios like searching through large photo albums, finding specific information across the internet, or monitoring environmental changes through satellite imagery. This paper explores the task of Multi-Image Visual Question Answering (MIQA): given a large set of images and a natural language query, the task is to generate a relevant and grounded response. We propose a new public benchmark, dubbed "Visual Haystacks (VHs)," specifically designed to evaluate LMMs' capabilities in visual retrieval and reasoning over sets of unrelated images, where we perform comprehensive evaluations demonstrating that even robust closed-source models struggle significantly. Towards addressing these shortcomings, we introduce MIRAGE (Multi-Image Retrieval Augmented Generation), a novel retrieval/QA framework tailored for LMMs that confronts the challenges of MIQA with marked efficiency and accuracy improvements over baseline methods. Our evaluation shows that MIRAGE surpasses closed-source GPT-4o models by up to 11% on the VHs benchmark and offers up to 3.4x improvements in efficiency over text-focused multi-stage approaches.

xT: Nested Tokenization for Larger Context in Large Images

Modern computer vision pipelines handle large images in one of two sub-optimal ways: down-sampling or cropping. These two methods incur significant losses in the amount of information and context present in an image. There are many downstream applications in which global context matters as much as high frequency details, such as in real-world satellite imagery; in such cases researchers have to make the uncomfortable choice of which information to discard. We introduce xT, a simple framework for vision transformers which effectively aggregates global context with local details and can model large images end-to-end on contemporary GPUs. We select a set of benchmark datasets across classic vision tasks which accurately reflect a vision model's ability to understand truly large images and incorporate fine details over large scales and assess our method's improvement on them. By introducing a nested tokenization scheme for large images in conjunction with long-sequence length models normally used for natural language processing, we are able to increase accuracy by up to 8.6% on challenging classification tasks and $F_1$ score by 11.6 on context-dependent segmentation in large images.

See, Say, and Segment: Teaching LMMs to Overcome False Premises

Current open-source Large Multimodal Models (LMMs) excel at tasks such as open-vocabulary language grounding and segmentation but can suffer under false premises when queries imply the existence of something that is not actually present in the image. We observe that existing methods that fine-tune an LMM to segment images significantly degrade their ability to reliably determine ("see") if an object is present and to interact naturally with humans ("say"), a form of catastrophic forgetting. In this work, we propose a cascading and joint training approach for LMMs to solve this task, avoiding catastrophic forgetting of previous skills. Our resulting model can "see" by detecting whether objects are present in an image, "say" by telling the user if they are not, proposing alternative queries or correcting semantic errors in the query, and finally "segment" by outputting the mask of the desired objects if they exist. Additionally, we introduce a novel False Premise Correction benchmark dataset, an extension of existing RefCOCO(+/g) referring segmentation datasets (which we call FP-RefCOCO(+/g)). The results show that our method not only detects false premises up to 55% better than existing approaches, but under false premise conditions produces relative cIOU improvements of more than 31% over baselines, and produces natural language feedback judged helpful up to 67% of the time.

ClimSim: An open large-scale dataset for training high-resolution physics emulators in hybrid multi-scale climate simulators

Modern climate projections lack adequate spatial and temporal resolution due to computational constraints. A consequence is inaccurate and imprecise prediction of critical processes such as storms. Hybrid methods that combine physics with machine learning (ML) have introduced a new generation of higher fidelity climate simulators that can sidestep Moore's Law by outsourcing compute-hungry, short, high-resolution simulations to ML emulators. However, this hybrid ML-physics simulation approach requires domain-specific treatment and has been inaccessible to ML experts because of lack of training data and relevant, easy-to-use workflows. We present ClimSim, the largest-ever dataset designed for hybrid ML-physics research. It comprises multi-scale climate simulations, developed by a consortium of climate scientists and ML researchers. It consists of 5.7 billion pairs of multivariate input and output vectors that isolate the influence of locally-nested, high-resolution, high-fidelity physics on a host climate simulator's macro-scale physical state. The dataset is global in coverage, spans multiple years at high sampling frequency, and is designed such that resulting emulators are compatible with downstream coupling into operational climate simulators. We implement a range of deterministic and stochastic regression baselines to highlight the ML challenges and their scoring. The data (https://huggingface.co/datasets/LEAP/ClimSim_high-res) and code (https://leap-stc.github.io/ClimSim) are released openly to support the development of hybrid ML-physics and high-fidelity climate simulations for the benefit of science and society.

Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learning

Remote sensing imagery provides comprehensive views of the Earth, where different sensors collect complementary data at different spatial scales. Large, pretrained models are commonly finetuned with imagery that is heavily augmented to mimic different conditions and scales, with the resulting models used for various tasks with imagery from a range of spatial scales. Such models overlook scale-specific information in the data. In this paper, we present Scale-MAE, a pretraining method that explicitly learns relationships between data at different, known scales throughout the pretraining process. Scale-MAE pretrains a network by masking an input image at a known input scale, where the area of the Earth covered by the image determines the scale of the ViT positional encoding, not the image resolution. Scale-MAE encodes the masked image with a standard ViT backbone, and then decodes the masked image through a bandpass filter to reconstruct low/high frequency images at lower/higher scales. We find that tasking the network with reconstructing both low/high frequency images leads to robust multiscale representations for remote sensing imagery. Scale-MAE achieves an average of a $5.0\%$ non-parametric kNN classification improvement across eight remote sensing datasets compared to current state-of-the-art and obtains a $0.9$ mIoU to $3.8$ mIoU improvement on the SpaceNet building segmentation transfer task for a range of evaluation scales.

Satlas: A Large-Scale, Multi-Task Dataset for Remote Sensing Image Understanding

Remote sensing images are useful for a wide variety of environmental and earth monitoring tasks, including tracking deforestation, illegal fishing, urban expansion, and natural disasters. The earth is extremely diverse -- the amount of potential tasks in remote sensing images is massive, and the sizes of features range from several kilometers to just tens of centimeters. However, creating generalizable computer vision methods is a challenge in part due to the lack of a large-scale dataset that captures these diverse features for many tasks. In this paper, we present Satlas, a remote sensing dataset and benchmark that is large in both breadth, featuring all of the aforementioned applications and more, as well as scale, comprising 290M labels under 137 categories and seven label modalities. We evaluate eight baselines and a proposed method on Satlas, and find that there is substantial room for improvement in addressing research challenges specific to remote sensing, including processing image time series that consist of images from very different types of sensors, and taking advantage of long-range spatial context. We also find that pre-training on Satlas substantially improves performance on downstream tasks with few labeled examples, increasing average accuracy by 16% over ImageNet and 5% over the next best baseline.

Snowpack Estimation in Key Mountainous Water Basins from Openly-Available, Multimodal Data Sources

Accurately estimating the snowpack in key mountainous basins is critical for water resource managers to make decisions that impact local and global economies, wildlife, and public policy. Currently, this estimation requires multiple LiDAR-equipped plane flights or in situ measurements, both of which are expensive, sparse, and biased towards accessible regions. In this paper, we demonstrate that fusing spatial and temporal information from multiple, openly-available satellite and weather data sources enables estimation of snowpack in key mountainous regions. Our multisource model outperforms single-source estimation by 5.0 inches RMSE, as well as outperforms sparse in situ measurements by 1.2 inches RMSE.

xView3-SAR: Detecting Dark Fishing Activity Using Synthetic Aperture Imagery

Unsustainable fishing practices worldwide pose a major threat to marine resources and ecosystems. Identifying vessels that evade monitoring systems -- known as "dark vessels" -- is key to managing and securing the health of marine environments. With the rise of satellite-based synthetic aperture radar (SAR) imaging and modern machine learning (ML), it is now possible to automate detection of dark vessels day or night, under all-weather conditions. SAR images, however, require domain-specific treatment and is not widely accessible to the ML community. Moreover, the objects (vessels) are small and sparse, challenging traditional computer vision approaches. We present the largest labeled dataset for training ML models to detect and characterize vessels from SAR. xView3-SAR consists of nearly 1,000 analysis-ready SAR images from the Sentinel-1 mission that are, on average, 29,400-by-24,400 pixels each. The images are annotated using a combination of automated and manual analysis. Co-located bathymetry and wind state rasters accompany every SAR image. We provide an overview of the results from the xView3 Computer Vision Challenge, an international competition using xView3-SAR for ship detection and characterization at large scale. We release the data (https://iuu.xview.us/) and code (https://github.com/DIUx-xView) to support ongoing development and evaluation of ML approaches for this important application.

WiSoSuper: Benchmarking Super-Resolution Methods on Wind and Solar Data

The transition to green energy grids depends on detailed wind and solar forecasts to optimize the siting and scheduling of renewable energy generation. Operational forecasts from numerical weather prediction models, however, only have a spatial resolution of 10 to 20-km, which leads to sub-optimal usage and development of renewable energy farms. Weather scientists have been developing super-resolution methods to increase the resolution, but often rely on simple interpolation techniques or computationally expensive differential equation-based models. Recently, machine learning-based models, specifically the physics-informed resolution-enhancing generative adversarial network (PhIREGAN), have outperformed traditional downscaling methods. We provide a thorough and extensible benchmark of leading deep learning-based super-resolution techniques, including the enhanced super-resolution generative adversarial network (ESRGAN) and an enhanced deep super-resolution (EDSR) network, on wind and solar data. We accompany the benchmark with a novel public, processed, and machine learning-ready dataset for benchmarking super-resolution methods on wind and solar data.

Proceedings of NeurIPS 2020 Workshop on Artificial Intelligence for Humanitarian Assistance and Disaster Response

These are the "proceedings" of the 2nd AI + HADR workshop which was held virtually on December 12, 2020 as part of the Neural Information Processing Systems conference. These are non-archival and merely serve as a way to collate all the papers accepted to the workshop.