Gregory Shakhnarovich

Takuma Yoneda, Tianchong Jiang, Gregory Shakhnarovich et al.

A core capability for robot manipulation is reasoning over where and how to stably place objects in cluttered environments. Traditionally, robots have relied on object-specific, hand-crafted heuristics in order to perform such reasoning, with limited generalizability beyond a small number of object instances and object interaction patterns. Recent approaches instead learn notions of physical interaction, namely motion prediction, but require supervision in the form of labeled object information or come at the cost of high sample complexity, and do not directly reason over stability or object placement. We present 6-DoFusion, a generative model capable of generating 3D poses of an object that produces a stable configuration of a given scene. Underlying 6-DoFusion is a diffusion model that incrementally refines a randomly initialized SE(3) pose to generate a sample from a learned, context-dependent distribution over stable poses. We evaluate our model on different object placement and stacking tasks, demonstrating its ability to construct stable scenes that involve novel object classes as well as to improve the accuracy of state-of-the-art 3D pose estimation methods.

Sunnie S. Y. Kim, Nicholas Kolkin, Jason Salavon et al.

Both geometry and texture are fundamental aspects of visual style. Existing style transfer methods, however, primarily focus on texture, almost entirely ignoring geometry. We propose deformable style transfer (DST), an optimization-based approach that jointly stylizes the texture and geometry of a content image to better match a style image. Unlike previous geometry-aware stylization methods, our approach is neither restricted to a particular domain (such as human faces), nor does it require training sets of matching style/content pairs. We demonstrate our method on a diverse set of content and style images including portraits, animals, objects, scenes, and paintings. Code has been made publicly available at https://github.com/sunniesuhyoung/DST.

Jiading Fang, Xiangshan Tan, Shengjie Lin et al.

If robots are to work effectively alongside people, they must be able to interpret natural language references to objects in their 3D environment. Understanding 3D referring expressions is challenging -- it requires the ability to both parse the 3D structure of the scene and correctly ground free-form language in the presence of distraction and clutter. We introduce Transcrib3D, an approach that brings together 3D detection methods and the emergent reasoning capabilities of large language models (LLMs). Transcrib3D uses text as the unifying medium, which allows us to sidestep the need to learn shared representations connecting multi-modal inputs, which would require massive amounts of annotated 3D data. As a demonstration of its effectiveness, Transcrib3D achieves state-of-the-art results on 3D reference resolution benchmarks, with a great leap in performance from previous multi-modality baselines. To improve upon zero-shot performance and facilitate local deployment on edge computers and robots, we propose self-correction for fine-tuning that trains smaller models, resulting in performance close to that of large models. We show that our method enables a real robot to perform pick-and-place tasks given queries that contain challenging referring expressions. Project site is at https://ripl.github.io/Transcrib3D.

Marcelo Sandoval-Castaneda, Bryan Russell, Josef Sivic et al.

Automated tools for video editing and assembly have applications ranging from filmmaking and advertisement to content creation for social media. Previous video editing work has mainly focused on either retrieval or user interfaces, leaving actual editing to the user. In contrast, we propose to automate the core task of video editing, formulating it as sequential decision making process. Ours is a multi-agent approach. We design an Editor agent and a Critic agent. The Editor takes as input a collection of video clips together with natural language instructions and uses tools commonly found in video editing software to produce an edited sequence. On the other hand, the Critic gives natural language feedback to the editor based on the produced sequence or renders it if it is satisfactory. We introduce a learning-based approach for enabling effective communication across specialized agents to address the language-driven video editing task. Finally, we explore an LLM-as-a-judge metric for evaluating the quality of video editing system and compare it with general human preference. We evaluate our system's output video sequences qualitatively and quantitatively through a user study and find that our system vastly outperforms existing approaches in terms of coverage, time constraint satisfaction, and human preference.

Marcelo Sandoval-Castaneda, Yanhong Li, Diane Brentari et al.

This paper presents an in-depth analysis of various self-supervision methods for isolated sign language recognition (ISLR). We consider four recently introduced transformer-based approaches to self-supervised learning from videos, and four pre-training data regimes, and study all the combinations on the WLASL2000 dataset. Our findings reveal that MaskFeat achieves performance superior to pose-based and supervised video models, with a top-1 accuracy of 79.02% on gloss-based WLASL2000. Furthermore, we analyze these models' ability to produce representations of ASL signs using linear probing on diverse phonological features. This study underscores the value of architecture and pre-training task choices in ISLR. Specifically, our results on WLASL2000 highlight the power of masked reconstruction pre-training, and our linear probing results demonstrate the importance of hierarchical vision transformers for sign language representation.

Jiading Fang, Shengjie Lin, Igor Vasiljevic et al.

A practical benefit of implicit visual representations like Neural Radiance Fields (NeRFs) is their memory efficiency: large scenes can be efficiently stored and shared as small neural nets instead of collections of images. However, operating on these implicit visual data structures requires extending classical image-based vision techniques (e.g., registration, blending) from image sets to neural fields. Towards this goal, we propose NeRFuser, a novel architecture for NeRF registration and blending that assumes only access to pre-generated NeRFs, and not the potentially large sets of images used to generate them. We propose registration from re-rendering, a technique to infer the transformation between NeRFs based on images synthesized from individual NeRFs. For blending, we propose sample-based inverse distance weighting to blend visual information at the ray-sample level. We evaluate NeRFuser on public benchmarks and a self-collected object-centric indoor dataset, showing the robustness of our method, including to views that are challenging to render from the individual source NeRFs.

Yuval Bahat, Gregory Shakhnarovich

Machine learning plays an increasingly significant role in many aspects of our lives (including medicine, transportation, security, justice and other domains), making the potential consequences of false predictions increasingly devastating. These consequences may be mitigated if we can automatically flag such false predictions and potentially assign them to alternative, more reliable mechanisms, that are possibly more costly and involve human attention. This suggests the task of detecting errors, which we tackle in this paper for the case of visual classification. To this end, we propose a novel approach for classification confidence estimation. We apply a set of semantics-preserving image transformations to the input image, and show how the resulting image sets can be used to estimate confidence in the classifier's prediction. We demonstrate the potential of our approach by extensively evaluating it on a wide variety of classifier architectures and datasets, including ResNext/ImageNet, achieving state of the art performance. This paper constitutes a significant revision of our earlier work in this direction (Bahat & Shakhnarovich, 2018).

Ruotian Luo, Gregory Shakhnarovich

We investigate the effect of different model architectures, training objectives, hyperparameter settings and decoding procedures on the diversity of automatically generated image captions. Our results show that 1) simple decoding by naive sampling, coupled with low temperature is a competitive and fast method to produce diverse and accurate caption sets; 2) training with CIDEr-based reward using Reinforcement learning harms the diversity properties of the resulting generator, which cannot be mitigated by manipulating decoding parameters. In addition, we propose a new metric AllSPICE for evaluating both accuracy and diversity of a set of captions by a single value.

Igor Vasiljevic, Nick Kolkin, Shanyi Zhang et al.

We introduce DIODE, a dataset that contains thousands of diverse high resolution color images with accurate, dense, long-range depth measurements. DIODE (Dense Indoor/Outdoor DEpth) is the first public dataset to include RGBD images of indoor and outdoor scenes obtained with one sensor suite. This is in contrast to existing datasets that focus on just one domain/scene type and employ different sensors, making generalization across domains difficult. The dataset is available for download at http://diode-dataset.org

Yuval Bahat, Michal Irani, Gregory Shakhnarovich

We propose an approach to distinguish between correct and incorrect image classifications. Our approach can detect misclassifications which either occur $\it{unintentionally}$ ("natural errors"), or due to $\it{intentional~adversarial~attacks}$ ("adversarial errors"), both in a single $\it{unified~framework}$. Our approach is based on the observation that correctly classified images tend to exhibit robust and consistent classifications under certain image transformations (e.g., horizontal flip, small image translation, etc.). In contrast, incorrectly classified images (whether due to adversarial errors or natural errors) tend to exhibit large variations in classification results under such transformations. Our approach does not require any modifications or retraining of the classifier, hence can be applied to any pre-trained classifier. We further use state of the art targeted adversarial attacks to demonstrate that even when the adversary has full knowledge of our method, the adversarial distortion needed for bypassing our detector is $\it{no~longer~imperceptible~to~the~human~eye}$. Our approach obtains state-of-the-art results compared to previous adversarial detection methods, surpassing them by a large margin.

Mohammadreza Mostajabi, Michael Maire, Gregory Shakhnarovich

We construct custom regularization functions for use in supervised training of deep neural networks. Our technique is applicable when the ground-truth labels themselves exhibit internal structure; we derive a regularizer by learning an autoencoder over the set of annotations. Training thereby becomes a two-phase procedure. The first phase models labels with an autoencoder. The second phase trains the actual network of interest by attaching an auxiliary branch that must predict output via a hidden layer of the autoencoder. After training, we discard this auxiliary branch. We experiment in the context of semantic segmentation, demonstrating this regularization strategy leads to consistent accuracy boosts over baselines, both when training from scratch, or in combination with ImageNet pretraining. Gains are also consistent over different choices of convolutional network architecture. As our regularizer is discarded after training, our method has zero cost at test time; the performance improvements are essentially free. We are simply able to learn better network weights by building an abstract model of the label space, and then training the network to understand this abstraction alongside the original task.

Yuval Bahat, Gregory Shakhnarovich

We develop a technique for automatically detecting the classification errors of a pre-trained visual classifier. Our method is agnostic to the form of the classifier, requiring access only to classifier responses to a set of inputs. We train a parametric binary classifier (error/correct) on a representation derived from a set of classifier responses generated from multiple copies of the same input, each subject to a different natural image transformation. Thus, we establish a measure of confidence in classifier's decision by analyzing the invariance of its decision under various transformations. In experiments with multiple data sets (STL-10,CIFAR-100,ImageNet) and classifiers, we demonstrate new state of the art for the error detection task. In addition, we apply our technique to novelty detection scenarios, where we also demonstrate state of the art results.

Ruotian Luo, Brian Price, Scott Cohen et al.

One property that remains lacking in image captions generated by contemporary methods is discriminability: being able to tell two images apart given the caption for one of them. We propose a way to improve this aspect of caption generation. By incorporating into the captioning training objective a loss component directly related to ability (by a machine) to disambiguate image/caption matches, we obtain systems that produce much more discriminative caption, according to human evaluation. Remarkably, our approach leads to improvement in other aspects of generated captions, reflected by a battery of standard scores such as BLEU, SPICE etc. Our approach is modular and can be applied to a variety of model/loss combinations commonly proposed for image captioning.

Herman Kamper, Gregory Shakhnarovich, Karen Livescu

There is growing interest in models that can learn from unlabelled speech paired with visual context. This setting is relevant for low-resource speech processing, robotics, and human language acquisition research. Here we study how a visually grounded speech model, trained on images of scenes paired with spoken captions, captures aspects of semantics. We use an external image tagger to generate soft text labels from images, which serve as targets for a neural model that maps untranscribed speech to (semantic) keyword labels. We introduce a newly collected data set of human semantic relevance judgements and an associated task, semantic speech retrieval, where the goal is to search for spoken utterances that are semantically relevant to a given text query. Without seeing any text, the model trained on parallel speech and images achieves a precision of almost 60% on its top ten semantic retrievals. Compared to a supervised model trained on transcriptions, our model matches human judgements better by some measures, especially in retrieving non-verbatim semantic matches. We perform an extensive analysis of the model and its resulting representations.

Nicholas Kolkin, Gregory Shakhnarovich, Eli Shechtman

In many computer vision tasks, for example saliency prediction or semantic segmentation, the desired output is a foreground map that predicts pixels where some criteria is satisfied. Despite the inherently spatial nature of this task commonly used learning objectives do not incorporate the spatial relationships between misclassified pixels and the underlying ground truth. The Weighted F-measure, a recently proposed evaluation metric, does reweight errors spatially, and has been shown to closely correlate with human evaluation of quality, and stably rank predictions with respect to noisy ground truths (such as a sloppy human annotator might generate). However it suffers from computational complexity which makes it intractable as an optimization objective for gradient descent, which must be evaluated thousands or millions of times while learning a model's parameters. We propose a differentiable and efficient approximation of this metric. By incorporating spatial information into the objective we can use a simpler model than competing methods without sacrificing accuracy, resulting in faster inference speeds and alleviating the need for pre/post-processing. We match (or improve) performance on several tasks compared to prior state of the art by traditional metrics, and in many cases significantly improve performance by the weighted F-measure.

Herman Kamper, Shane Settle, Gregory Shakhnarovich et al.

During language acquisition, infants have the benefit of visual cues to ground spoken language. Robots similarly have access to audio and visual sensors. Recent work has shown that images and spoken captions can be mapped into a meaningful common space, allowing images to be retrieved using speech and vice versa. In this setting of images paired with untranscribed spoken captions, we consider whether computer vision systems can be used to obtain textual labels for the speech. Concretely, we use an image-to-words multi-label visual classifier to tag images with soft textual labels, and then train a neural network to map from the speech to these soft targets. We show that the resulting speech system is able to predict which words occur in an utterance---acting as a spoken bag-of-words classifier---without seeing any parallel speech and text. We find that the model often confuses semantically related words, e.g. "man" and "person", making it even more effective as a semantic keyword spotter.

Gustav Larsson, Michael Maire, Gregory Shakhnarovich

We investigate and improve self-supervision as a drop-in replacement for ImageNet pretraining, focusing on automatic colorization as the proxy task. Self-supervised training has been shown to be more promising for utilizing unlabeled data than other, traditional unsupervised learning methods. We build on this success and evaluate the ability of our self-supervised network in several contexts. On VOC segmentation and classification tasks, we present results that are state-of-the-art among methods not using ImageNet labels for pretraining representations. Moreover, we present the first in-depth analysis of self-supervision via colorization, concluding that formulation of the loss, training details and network architecture play important roles in its effectiveness. This investigation is further expanded by revisiting the ImageNet pretraining paradigm, asking questions such as: How much training data is needed? How many labels are needed? How much do features change when fine-tuned? We relate these questions back to self-supervision by showing that colorization provides a similarly powerful supervisory signal as various flavors of ImageNet pretraining.

Ruotian Luo, Gregory Shakhnarovich

We consider generation and comprehension of natural language referring expression for objects in an image. Unlike generic "image captioning" which lacks natural standard evaluation criteria, quality of a referring expression may be measured by the receiver's ability to correctly infer which object is being described. Following this intuition, we propose two approaches to utilize models trained for comprehension task to generate better expressions. First, we use a comprehension module trained on human-generated expressions, as a "critic" of referring expression generator. The comprehension module serves as a differentiable proxy of human evaluation, providing training signal to the generation module. Second, we use the comprehension module in a generate-and-rerank pipeline, which chooses from candidate expressions generated by a model according to their performance on the comprehension task. We show that both approaches lead to improved referring expression generation on multiple benchmark datasets.

Mohammadreza Mostajabi, Nicholas Kolkin, Gregory Shakhnarovich

We propose an approach for learning category-level semantic segmentation purely from image-level classification tags indicating presence of categories. It exploits localization cues that emerge from training classification-tasked convolutional networks, to drive a "self-supervision" process that automatically labels a sparse, diverse training set of points likely to belong to classes of interest. Our approach has almost no hyperparameters, is modular, and allows for very fast training of segmentation in less than 3 minutes. It obtains competitive results on the VOC 2012 segmentation benchmark. More, significantly the modularity and fast training of our framework allows new classes to efficiently added for inference.

Igor Vasiljevic, Ayan Chakrabarti, Gregory Shakhnarovich

State-of-the-art algorithms for many semantic visual tasks are based on the use of convolutional neural networks. These networks are commonly trained, and evaluated, on large annotated datasets of artifact-free high-quality images. In this paper, we investigate the effect of one such artifact that is quite common in natural capture settings: optical blur. We show that standard network models, trained only on high-quality images, suffer a significant degradation in performance when applied to those degraded by blur due to defocus, or subject or camera motion. We investigate the extent to which this degradation is due to the mismatch between training and input image statistics. Specifically, we find that fine-tuning a pre-trained model with blurred images added to the training set allows it to regain much of the lost accuracy. We also show that there is a fair amount of generalization between different degrees and types of blur, which implies that a single network model can be used robustly for recognition when the nature of the blur in the input is unknown. We find that this robustness arises as a result of these models learning to generate blur invariant representations in their hidden layers. Our findings provide useful insights towards developing vision systems that can perform reliably on real world images affected by blur.

Taehwan Kim, Jonathan Keane, Weiran Wang et al.

We study the problem of recognizing video sequences of fingerspelled letters in American Sign Language (ASL). Fingerspelling comprises a significant but relatively understudied part of ASL. Recognizing fingerspelling is challenging for a number of reasons: It involves quick, small motions that are often highly coarticulated; it exhibits significant variation between signers; and there has been a dearth of continuous fingerspelling data collected. In this work we collect and annotate a new data set of continuous fingerspelling videos, compare several types of recognizers, and explore the problem of signer variation. Our best-performing models are segmental (semi-Markov) conditional random fields using deep neural network-based features. In the signer-dependent setting, our recognizers achieve up to about 92% letter accuracy. The multi-signer setting is much more challenging, but with neural network adaptation we achieve up to 83% letter accuracies in this setting.

Gustav Larsson, Michael Maire, Gregory Shakhnarovich

We introduce a design strategy for neural network macro-architecture based on self-similarity. Repeated application of a simple expansion rule generates deep networks whose structural layouts are precisely truncated fractals. These networks contain interacting subpaths of different lengths, but do not include any pass-through or residual connections; every internal signal is transformed by a filter and nonlinearity before being seen by subsequent layers. In experiments, fractal networks match the excellent performance of standard residual networks on both CIFAR and ImageNet classification tasks, thereby demonstrating that residual representations may not be fundamental to the success of extremely deep convolutional neural networks. Rather, the key may be the ability to transition, during training, from effectively shallow to deep. We note similarities with student-teacher behavior and develop drop-path, a natural extension of dropout, to regularize co-adaptation of subpaths in fractal architectures. Such regularization allows extraction of high-performance fixed-depth subnetworks. Additionally, fractal networks exhibit an anytime property: shallow subnetworks provide a quick answer, while deeper subnetworks, with higher latency, provide a more accurate answer.

Ayan Chakrabarti, Jingyu Shao, Gregory Shakhnarovich

A single color image can contain many cues informative towards different aspects of local geometric structure. We approach the problem of monocular depth estimation by using a neural network to produce a mid-level representation that summarizes these cues. This network is trained to characterize local scene geometry by predicting, at every image location, depth derivatives of different orders, orientations and scales. However, instead of a single estimate for each derivative, the network outputs probability distributions that allow it to express confidence about some coefficients, and ambiguity about others. Scene depth is then estimated by harmonizing this overcomplete set of network predictions, using a globalization procedure that finds a single consistent depth map that best matches all the local derivative distributions. We demonstrate the efficacy of this approach through evaluation on the NYU v2 depth data set.

Gustav Larsson, Michael Maire, Gregory Shakhnarovich

We develop a fully automatic image colorization system. Our approach leverages recent advances in deep networks, exploiting both low-level and semantic representations. As many scene elements naturally appear according to multimodal color distributions, we train our model to predict per-pixel color histograms. This intermediate output can be used to automatically generate a color image, or further manipulated prior to image formation. On both fully and partially automatic colorization tasks, we outperform existing methods. We also explore colorization as a vehicle for self-supervised visual representation learning.

Mohammadreza Mostajabi, Payman Yadollahpour, Gregory Shakhnarovich

We introduce a purely feed-forward architecture for semantic segmentation. We map small image elements (superpixels) to rich feature representations extracted from a sequence of nested regions of increasing extent. These regions are obtained by "zooming out" from the superpixel all the way to scene-level resolution. This approach exploits statistical structure in the image and in the label space without setting up explicit structured prediction mechanisms, and thus avoids complex and expensive inference. Instead superpixels are classified by a feedforward multilayer network. Our architecture achieves new state of the art performance in semantic segmentation, obtaining 64.4% average accuracy on the PASCAL VOC 2012 test set.