Pradeep Sen

Chengyuan Xu, Boning Dong, Noah Stier et al.

We introduce an interactive image segmentation and visualization framework for identifying, inspecting, and editing tiny objects (just a few pixels wide) in large multi-megapixel high-dynamic-range (HDR) images. Detecting cosmic rays (CRs) in astronomical observations is a cumbersome workflow that requires multiple tools, so we developed an interactive toolkit that unifies model inference, HDR image visualization, segmentation mask inspection and editing into a single graphical user interface. The feature set, initially designed for astronomical data, makes this work a useful research-supporting tool for human-in-the-loop tiny-object segmentation in scientific areas like biomedicine, materials science, remote sensing, etc., as well as computer vision. Our interface features mouse-controlled, synchronized, dual-window visualization of the image and the segmentation mask, a critical feature for locating tiny objects in multi-megapixel images. The browser-based tool can be readily hosted on the web to provide multi-user access and GPU acceleration for any device. The toolkit can also be used as a high-precision annotation tool, or adapted as the frontend for an interactive machine learning framework. Our open-source dataset, CR detection model, and visualization toolkit are available at https://github.com/cy-xu/cosmic-conn.

Sherry X. Chen, Yaron Vaxman, Elad Ben Baruch et al.

We propose Image Content Appeal Assessment (ICAA), a novel metric that quantifies the level of positive interest an image's content generates for viewers, such as the appeal of food in a photograph. This is fundamentally different from traditional Image-Aesthetics Assessment (IAA), which judges an image's artistic quality. While previous studies often confuse the concepts of ``aesthetics'' and ``appeal,'' our work addresses this by being the first to study ICAA explicitly. To do this, we propose a novel system that automates dataset creation and implements algorithms to estimate and boost content appeal. We use our pipeline to generate two large-scale datasets (70K+ images each) in diverse domains (food and room interior design) to train our models, which revealed little correlation between content appeal and aesthetics. Our user study, with more than 76% of participants preferring the appeal-enhanced images, confirms that our appeal ratings accurately reflect user preferences, establishing ICAA as a unique evaluative criterion. Our code and datasets are available at https://github.com/SherryXTChen/AID-Appeal.

Steve Bako, Pradeep Sen, Anton Kaplanyan

Physically based rendering of complex scenes can be prohibitively costly with a potentially unbounded and uneven distribution of complexity across the rendered image. The goal of an ideal level of detail (LoD) method is to make rendering costs independent of the 3D scene complexity, while preserving the appearance of the scene. However, current prefiltering LoD methods are limited in the appearances they can support due to their reliance of approximate models and other heuristics. We propose the first comprehensive multi-scale LoD framework for prefiltering 3D environments with complex geometry and materials (e.g., the Disney BRDF), while maintaining the appearance with respect to the ray-traced reference. Using a multi-scale hierarchy of the scene, we perform a data-driven prefiltering step to obtain an appearance phase function and directional coverage mask at each scale. At the heart of our approach is a novel neural representation that encodes this information into a compact latent form that is easy to decode inside a physically based renderer. Once a scene is baked out, our method requires no original geometry, materials, or textures at render time. We demonstrate that our approach compares favorably to state-of-the-art prefiltering methods and achieves considerable savings in memory for complex scenes.

Sherry X. Chen, Alex Christopher Lim, Yimeng Liu et al.

Virtual try-on (VTO) applications aim to replicate the in-store shopping experience and enhance online shopping by enabling users to interact with garments. However, many existing tools adopt a one-size-fits-all approach when visualizing clothing items. This approach limits user interaction with garments, particularly regarding size and fit adjustments, and fails to provide direct insights for size recommendations. As a result, these limitations contribute to high return rates in online shopping. To address this, we introduce SiCo, a new online VTO system that allows users to upload images of themselves and interact with garments by visualizing how different sizes would fit their bodies. Our user study demonstrates that our approach significantly improves users' ability to assess how outfits will appear on their bodies and increases their confidence in selecting clothing sizes that align with their preferences. Based on our evaluation, we believe that SiCo has the potential to reduce return rates and transform the online clothing shopping experience.

Sherry X. Chen, Yaron Vaxman, Elad Ben Baruch et al.

Despite many attempts to leverage pre-trained text-to-image models (T2I) like Stable Diffusion (SD) for controllable image editing, producing good predictable results remains a challenge. Previous approaches have focused on either fine-tuning pre-trained T2I models on specific datasets to generate certain kinds of images (e.g., with a specific object or person), or on optimizing the weights, text prompts, and/or learning features for each input image in an attempt to coax the image generator to produce the desired result. However, these approaches all have shortcomings and fail to produce good results in a predictable and controllable manner. To address this problem, we present TiNO-Edit, an SD-based method that focuses on optimizing the noise patterns and diffusion timesteps during editing, something previously unexplored in the literature. With this simple change, we are able to generate results that both better align with the original images and reflect the desired result. Furthermore, we propose a set of new loss functions that operate in the latent domain of SD, greatly speeding up the optimization when compared to prior approaches, which operate in the pixel domain. Our method can be easily applied to variations of SD including Textual Inversion and DreamBooth that encode new concepts and incorporate them into the edited results. We present a host of image-editing capabilities enabled by our approach. Our code is publicly available at https://github.com/SherryXTChen/TiNO-Edit.

Sherry X. Chen, Misha Sra, Pradeep Sen

Although natural language instructions offer an intuitive way to guide automated image editing, deep-learning models often struggle to achieve high-quality results, largely due to the difficulty of creating large, high-quality training datasets. To do this, previous approaches have typically relied on text-to-image (T2I) generative models to produce pairs of original and edited images that simulate the input/output of an instruction-guided image-editing model. However, these image pairs often fail to align with the specified edit instructions due to the limitations of T2I models, which negatively impacts models trained on such datasets. To address this, we present Instruct-CLIP (I-CLIP), a selfsupervised method that learns the semantic changes between original and edited images to refine and better align the instructions in existing datasets. Furthermore, we adapt Instruct-CLIP to handle noisy latent images and diffusion timesteps so that it can be used to train latent diffusion models (LDMs) and efficiently enforce alignment between the edit instruction and the image changes in latent space at any step of the diffusion pipeline. We use Instruct-CLIP to correct the InstructPix2Pix dataset and get over 120K refined samples we then use to fine-tune their model, guided by our novel I-CLIP-based loss function. The resulting model can produce edits that are more aligned with the given instructions. Our code and dataset are available at https://github.com/SherryXTChen/Instruct-CLIP.git.

Chengyuan Xu, Curtis McCully, Boning Dong et al.

Rejecting cosmic rays (CRs) is essential for the scientific interpretation of CCD-captured data, but detecting CRs in single-exposure images has remained challenging. Conventional CR detectors require experimental parameter tuning for different instruments, and recent deep learning methods only produce instrument-specific models that suffer from performance loss on telescopes not included in the training data. We present Cosmic-CoNN, a generic CR detector deployed for 24 telescopes at the Las Cumbres Observatory, which is made possible by the three contributions in this work: 1) We build a large and diverse ground-based CR dataset leveraging thousands of images from a global telescope network. 2) We propose a novel loss function and a neural network optimized for telescope imaging data to train generic CR detection models. At 95% recall, our model achieves a precision of 93.70% on Las Cumbres imaging data and maintains a consistent performance on new ground-based instruments never used for training. Specifically, the Cosmic-CoNN model trained on the Las Cumbres CR dataset maintains high precisions of 92.03% and 96.69% on Gemini GMOS-N/S 1x1 and 2x2 binning images, respectively. 3) We build a suite of tools including an interactive CR mask visualization and editing interface, console commands, and Python APIs to make automatic, robust CR detection widely accessible by the community of astronomers. Our dataset, open-source codebase, and trained models are available at https://github.com/cy-xu/cosmic-conn.

Ekta Prashnani, Orazio Gallo, Joohwan Kim et al.

We tackle the problem of predicting saliency maps for videos of dynamic scenes. We note that the accuracy of the maps reconstructed from the gaze data of a fixed number of observers varies with the frame, as it depends on the content of the scene. This issue is particularly pressing when a limited number of observers are available. In such cases, directly minimizing the discrepancy between the predicted and measured saliency maps, as traditional deep-learning methods do, results in overfitting to the noisy data. We propose a noise-aware training (NAT) paradigm that quantifies and accounts for the uncertainty arising from frame-specific gaze data inaccuracy. We show that NAT is especially advantageous when limited training data is available, with experiments across different models, loss functions, and datasets. We also introduce a video game-based saliency dataset, with rich temporal semantics, and multiple gaze attractors per frame. The dataset and source code are available at https://github.com/NVlabs/NAT-saliency.

Yang Zhou, Tao Huang, Ravi Ramamoorthi et al.

Creating an appearance-preserving level-of-detail (LoD) representation for arbitrary 3D scenes is a challenging problem. The appearance of a scene is an intricate combination of both geometry and material models, and is further complicated by correlation due to the spatial configuration of scene elements. We present a novel volumetric representation for the aggregated appearance of complex scenes and an efficient pipeline for LoD generation and rendering. The core of our representation is the Aggregated Bidirectional Scattering Distribution Function (ABSDF) that summarizes the far-field appearance of all surfaces inside a voxel. We propose a closed-form factorization of the ABSDF that accounts for spatially varying and orientation-varying material parameters. We tackle the challenge of capturing the correlation existing locally within a voxel and globally across different parts of the scene. Our method faithfully reproduces appearance and achieves higher quality than existing scene filtering methods while being inherently efficient to render. The memory footprint and rendering cost of our representation are independent of the original scene complexity.

Alex Rich, Noah Stier, Pradeep Sen et al.

The promise of unsupervised multi-view-stereo (MVS) is to leverage large unlabeled datasets, yet current methods underperform when training on difficult data, such as handheld smartphone videos of indoor scenes. Meanwhile, high-quality synthetic datasets are available but MVS networks trained on these datasets fail to generalize to real-world examples. To bridge this gap, we propose a semi-supervised learning framework that allows us to train on real and rendered images jointly, capturing structural priors from synthetic data while ensuring parity with the real-world domain. Central to our framework is a novel set of losses that leverages powerful existing monocular relative-depth estimators trained on the synthetic dataset, transferring the rich structure of this relative depth to the MVS predictions on unlabeled data. Inspired by perceptual image metrics, we compare the MVS and monocular predictions via a deep feature loss and a multi-scale statistical loss. Our full framework, which we call Prism, achieves large quantitative and qualitative improvements over current unsupervised and synthetic-supervised MVS networks. This is a best-case-scenario result, opening the door to using both unlabeled smartphone videos and photorealistic synthetic datasets for training MVS networks.

Noah Stier, Alexander Rich, Pradeep Sen et al.

Recent volumetric 3D reconstruction methods can produce very accurate results, with plausible geometry even for unobserved surfaces. However, they face an undesirable trade-off when it comes to multi-view fusion. They can fuse all available view information by global averaging, thus losing fine detail, or they can heuristically cluster views for local fusion, thus restricting their ability to consider all views jointly. Our key insight is that greater detail can be retained without restricting view diversity by learning a view-fusion function conditioned on camera pose and image content. We propose to learn this multi-view fusion using a transformer. To this end, we introduce VoRTX, an end-to-end volumetric 3D reconstruction network using transformers for wide-baseline, multi-view feature fusion. Our model is occlusion-aware, leveraging the transformer architecture to predict an initial, projective scene geometry estimate. This estimate is used to avoid backprojecting image features through surfaces into occluded regions. We train our model on ScanNet and show that it produces better reconstructions than state-of-the-art methods. We also demonstrate generalization without any fine-tuning, outperforming the same state-of-the-art methods on two other datasets, TUM-RGBD and ICL-NUIM.

Alexander Rich, Noah Stier, Pradeep Sen et al.

We present 3DVNet, a novel multi-view stereo (MVS) depth-prediction method that combines the advantages of previous depth-based and volumetric MVS approaches. Our key idea is the use of a 3D scene-modeling network that iteratively updates a set of coarse depth predictions, resulting in highly accurate predictions which agree on the underlying scene geometry. Unlike existing depth-prediction techniques, our method uses a volumetric 3D convolutional neural network (CNN) that operates in world space on all depth maps jointly. The network can therefore learn meaningful scene-level priors. Furthermore, unlike existing volumetric MVS techniques, our 3D CNN operates on a feature-augmented point cloud, allowing for effective aggregation of multi-view information and flexible iterative refinement of depth maps. Experimental results show our method exceeds state-of-the-art accuracy in both depth prediction and 3D reconstruction metrics on the ScanNet dataset, as well as a selection of scenes from the TUM-RGBD and ICL-NUIM datasets. This shows that our method is both effective and generalizes to new settings.

Yi Ding, Alex Rich, Mason Wang et al.

Multimodal classification is a core task in human-centric machine learning. We observe that information is highly complementary across modalities, thus unimodal information can be drastically sparsified prior to multimodal fusion without loss of accuracy. To this end, we present Sparse Fusion Transformers (SFT), a novel multimodal fusion method for transformers that performs comparably to existing state-of-the-art methods while having greatly reduced memory footprint and computation cost. Key to our idea is a sparse-pooling block that reduces unimodal token sets prior to cross-modality modeling. Evaluations are conducted on multiple multimodal benchmark datasets for a wide range of classification tasks. State-of-the-art performance is obtained on multiple benchmarks under similar experiment conditions, while reporting up to six-fold reduction in computational cost and memory requirements. Extensive ablation studies showcase our benefits of combining sparsification and multimodal learning over naive approaches. This paves the way for enabling multimodal learning on low-resource devices.

Abhishek Badki, Orazio Gallo, Jan Kautz et al.

Time-to-contact (TTC), the time for an object to collide with the observer's plane, is a powerful tool for path planning: it is potentially more informative than the depth, velocity, and acceleration of objects in the scene -- even for humans. TTC presents several advantages, including requiring only a monocular, uncalibrated camera. However, regressing TTC for each pixel is not straightforward, and most existing methods make over-simplifying assumptions about the scene. We address this challenge by estimating TTC via a series of simpler, binary classifications. We predict with low latency whether the observer will collide with an obstacle within a certain time, which is often more critical than knowing exact, per-pixel TTC. For such scenarios, our method offers a temporal geofence in 6.4 ms -- over 25x faster than existing methods. Our approach can also estimate per-pixel TTC with arbitrarily fine quantization (including continuous values), when the computational budget allows for it. To the best of our knowledge, our method is the first to offer TTC information (binary or coarsely quantized) at sufficiently high frame-rates for practical use.

Abhishek Badki, Alejandro Troccoli, Kihwan Kim et al.

Stereo-based depth estimation is a cornerstone of computer vision, with state-of-the-art methods delivering accurate results in real time. For several applications such as autonomous navigation, however, it may be useful to trade accuracy for lower latency. We present Bi3D, a method that estimates depth via a series of binary classifications. Rather than testing if objects are at a particular depth $D$, as existing stereo methods do, it classifies them as being closer or farther than $D$. This property offers a powerful mechanism to balance accuracy and latency. Given a strict time budget, Bi3D can detect objects closer than a given distance in as little as a few milliseconds, or estimate depth with arbitrarily coarse quantization, with complexity linear with the number of quantization levels. Bi3D can also use the allotted quantization levels to get continuous depth, but in a specific depth range. For standard stereo (i.e., continuous depth on the whole range), our method is close to or on par with state-of-the-art, finely tuned stereo methods.

Abhishek Badki, Orazio Gallo, Jan Kautz et al.

Estimating a mesh from an unordered set of sparse, noisy 3D points is a challenging problem that requires carefully selected priors. Existing hand-crafted priors, such as smoothness regularizers, impose an undesirable trade-off between attenuating noise and preserving local detail. Recent deep-learning approaches produce impressive results by learning priors directly from the data. However, the priors are learned at the object level, which makes these algorithms class-specific and even sensitive to the pose of the object. We introduce meshlets, small patches of mesh that we use to learn local shape priors. Meshlets act as a dictionary of local features and thus allow to use learned priors to reconstruct object meshes in any pose and from unseen classes, even when the noise is large and the samples sparse.

Ekta Prashnani, Hong Cai, Yasamin Mostofi et al.

The ability to estimate the perceptual error between images is an important problem in computer vision with many applications. Although it has been studied extensively, however, no method currently exists that can robustly predict visual differences like humans. Some previous approaches used hand-coded models, but they fail to model the complexity of the human visual system. Others used machine learning to train models on human-labeled datasets, but creating large, high-quality datasets is difficult because people are unable to assign consistent error labels to distorted images. In this paper, we present a new learning-based method that is the first to predict perceptual image error like human observers. Since it is much easier for people to compare two given images and identify the one more similar to a reference than to assign quality scores to each, we propose a new, large-scale dataset labeled with the probability that humans will prefer one image over another. We then train a deep-learning model using a novel, pairwise-learning framework to predict the preference of one distorted image over the other. Our key observation is that our trained network can then be used separately with only one distorted image and a reference to predict its perceptual error, without ever being trained on explicit human perceptual-error labels. The perceptual error estimated by our new metric, PieAPP, is well-correlated with human opinion. Furthermore, it significantly outperforms existing algorithms, beating the state-of-the-art by almost 3x on our test set in terms of binary error rate, while also generalizing to new kinds of distortions, unlike previous learning-based methods.

Chieh-Chi Kao, Yuxiang Wang, Jonathan Waltman et al.

Image hallucination and super-resolution have been studied for decades, and many approaches have been proposed to upsample low-resolution images using information from the images themselves, multiple example images, or large image databases. However, most of this work has focused exclusively on small magnification levels because the algorithms simply sharpen the blurry edges in the upsampled images - no actual new detail is typically reconstructed in the final result. In this paper, we present a patch-based algorithm for image hallucination which, for the first time, properly synthesizes novel high frequency detail. To do this, we pose the synthesis problem as a patch-based optimization which inserts coherent, high-frequency detail from contextually-similar images of the same physical scene/subject provided from either a personal image collection or a large online database. The resulting image is visually plausible and contains coherent high frequency information. We demonstrate the robustness of our algorithm by testing it on a large number of images and show that its performance is considerably superior to all state-of-the-art approaches, a result that is verified to be statistically significant through a randomized user study.

Shayan Sadigh, Pradeep Sen

We describe a new class of subsampling techniques for CNNs, termed multisampling, that significantly increases the amount of information kept by feature maps through subsampling layers. One version of our method, which we call checkered subsampling, significantly improves the accuracy of state-of-the-art architectures such as DenseNet and ResNet without any additional parameters and, remarkably, improves the accuracy of certain pretrained ImageNet models without any training or fine-tuning. We glean possible insight into the nature of data augmentations and demonstrate experimentally that coarse feature maps are bottlenecking the performance of neural networks in image classification.

Chieh-Chi Kao, Teng-Yok Lee, Pradeep Sen et al.

Active learning - a class of algorithms that iteratively searches for the most informative samples to include in a training dataset - has been shown to be effective at annotating data for image classification. However, the use of active learning for object detection is still largely unexplored as determining informativeness of an object-location hypothesis is more difficult. In this paper, we address this issue and present two metrics for measuring the informativeness of an object hypothesis, which allow us to leverage active learning to reduce the amount of annotated data needed to achieve a target object detection performance. Our first metric measures 'localization tightness' of an object hypothesis, which is based on the overlapping ratio between the region proposal and the final prediction. Our second metric measures 'localization stability' of an object hypothesis, which is based on the variation of predicted object locations when input images are corrupted by noise. Our experimental results show that by augmenting a conventional active-learning algorithm designed for classification with the proposed metrics, the amount of labeled training data required can be reduced up to 25%. Moreover, on PASCAL 2007 and 2012 datasets our localization-stability method has an average relative improvement of 96.5% and 81.9% over the baseline method using classification only.

Qiaodong Cui, Victor Fragoso, Chris Sweeney et al.

We present GraphMatch, an approximate yet efficient method for building the matching graph for large-scale structure-from-motion (SfM) pipelines. Unlike modern SfM pipelines that use vocabulary (Voc.) trees to quickly build the matching graph and avoid a costly brute-force search of matching image pairs, GraphMatch does not require an expensive offline pre-processing phase to construct a Voc. tree. Instead, GraphMatch leverages two priors that can predict which image pairs are likely to match, thereby making the matching process for SfM much more efficient. The first is a score computed from the distance between the Fisher vectors of any two images. The second prior is based on the graph distance between vertices in the underlying matching graph. GraphMatch combines these two priors into an iterative "sample-and-propagate" scheme similar to the PatchMatch algorithm. Its sampling stage uses Fisher similarity priors to guide the search for matching image pairs, while its propagation stage explores neighbors of matched pairs to find new ones with a high image similarity score. Our experiments show that GraphMatch finds the most image pairs as compared to competing, approximate methods while at the same time being the most efficient.

Victor Fragoso, Chris Sweeney, Pradeep Sen et al.

While RANSAC-based methods are robust to incorrect image correspondences (outliers), their hypothesis generators are not robust to correct image correspondences (inliers) with positional error (noise). This slows down their convergence because hypotheses drawn from a minimal set of noisy inliers can deviate significantly from the optimal model. This work addresses this problem by introducing ANSAC, a RANSAC-based estimator that accounts for noise by adaptively using more than the minimal number of correspondences required to generate a hypothesis. ANSAC estimates the inlier ratio (the fraction of correct correspondences) of several ranked subsets of candidate correspondences and generates hypotheses from them. Its hypothesis-generation mechanism prioritizes the use of subsets with high inlier ratio to generate high-quality hypotheses. ANSAC uses an early termination criterion that keeps track of the inlier ratio history and terminates when it has not changed significantly for a period of time. The experiments show that ANSAC finds good homography and fundamental matrix estimates in a few iterations, consistently outperforming state-of-the-art methods.