This paper explores the potential of leveraging language priors learned by text-to-image diffusion models to address ambiguity and visual nuisance in monocular depth estimation. Particularly, traditional monocular depth estimation suffers from inherent ambiguity due to the absence of stereo or multi-view depth cues, and nuisance due to lack of robustness of vision. We argue that language prior in diffusion models can enhance monocular depth estimation by leveraging the geometric prior aligned with the language description, which is learned during text-to-image pre-training. To generate images that reflect the text properly, the model must comprehend the size and shape of specified objects, their spatial relationship, and the scale of the scene. Thus, we propose PriorDiffusion, using a pre-trained text-to-image diffusion model that takes both image and text description that aligned with the scene to infer affine-invariant depth through a denoising process. We also show that language priors can guide the model's attention to specific regions and help it perceive the 3D scene in alignment with user intent. Simultaneously, it acts as a constraint to accelerate the convergence of the diffusion trajectory, since learning 3D properties from a condensed, low-dimensional language feature is more efficient compared with learning from a redundant, high-dimensional image feature. By training on HyperSim and Virtual KITTI, we achieve state-of-the-art zero-shot performance and a faster convergence speed, compared with other diffusion-based depth estimators, across NYUv2, KITTI, ETH3D, and ScanNet.

We propose UnCLe, a standardized benchmark for Unsupervised Continual Learning of a multimodal depth estimation task: Depth completion aims to infer a dense depth map from a pair of synchronized RGB image and sparse depth map. We benchmark depth completion models under the practical scenario of unsupervised learning over continuous streams of data. Existing methods are typically trained on a static, or stationary, dataset. However, when adapting to novel non-stationary distributions, they "catastrophically forget" previously learned information. UnCLe simulates these non-stationary distributions by adapting depth completion models to sequences of datasets containing diverse scenes captured from distinct domains using different visual and range sensors. We adopt representative methods from continual learning paradigms and translate them to enable unsupervised continual learning of depth completion. We benchmark these models for indoor and outdoor and investigate the degree of catastrophic forgetting through standard quantitative metrics. Furthermore, we introduce model inversion quality as an additional measure of forgetting. We find that unsupervised continual learning of depth completion is an open problem, and we invite researchers to leverage UnCLe as a development platform.

Three-dimensional (3D) reconstruction from a single image is an ill-posed problem with inherent ambiguities, i.e. scale. Predicting a 3D scene from text description(s) is similarly ill-posed, i.e. spatial arrangements of objects described. We investigate the question of whether two inherently ambiguous modalities can be used in conjunction to produce metric-scaled reconstructions. To test this, we focus on monocular depth estimation, the problem of predicting a dense depth map from a single image, but with an additional text caption describing the scene. To this end, we begin by encoding the text caption as a mean and standard deviation; using a variational framework, we learn the distribution of the plausible metric reconstructions of 3D scenes corresponding to the text captions as a prior. To "select" a specific reconstruction or depth map, we encode the given image through a conditional sampler that samples from the latent space of the variational text encoder, which is then decoded to the output depth map. Our approach is trained alternatingly between the text and image branches: in one optimization step, we predict the mean and standard deviation from the text description and sample from a standard Gaussian, and in the other, we sample using a (image) conditional sampler. Once trained, we directly predict depth from the encoded text using the conditional sampler. We demonstrate our approach on indoor (NYUv2) and outdoor (KITTI) scenarios, where we show that language can consistently improve performance in both.

We propose a method to infer semantic segmentation maps from images captured under adverse weather conditions. We begin by examining existing models on images degraded by weather conditions such as rain, fog, or snow, and found that they exhibit a large performance drop as compared to those captured under clear weather. To control for changes in scene structures, we propose WeatherProof, the first semantic segmentation dataset with accurate clear and adverse weather image pairs that share an underlying scene. Through this dataset, we analyze the error modes in existing models and found that they were sensitive to the highly complex combination of different weather effects induced on the image during capture. To improve robustness, we propose a way to use language as guidance by identifying contributions of adverse weather conditions and injecting that as "side information". Models trained using our language guidance exhibit performance gains by up to 10.2% in mIoU on WeatherProof, up to 8.44% in mIoU on the widely used ACDC dataset compared to standard training techniques, and up to 6.21% in mIoU on the ACDC dataset as compared to previous SOTA methods.

This report reviews the results of the GT-Rain challenge on single image deraining at the UG2+ workshop at CVPR 2023. The aim of this competition is to study the rainy weather phenomenon in real world scenarios, provide a novel real world rainy image dataset, and to spark innovative ideas that will further the development of single image deraining methods on real images. Submissions were trained on the GT-Rain dataset and evaluated on an extension of the dataset consisting of 15 additional scenes. Scenes in GT-Rain are comprised of real rainy image and ground truth image captured moments after the rain had stopped. 275 participants were registered in the challenge and 55 competed in the final testing phase.

It is common to observe performance degradation when transferring models trained on some (source) datasets to target testing data due to a domain gap between them. Existing methods for bridging this gap, such as domain adaptation (DA), may require the source data on which the model was trained (often not available), while others, i.e., source-free DA, require many passes through the testing data. We propose an online test-time adaptation method for depth completion, the task of inferring a dense depth map from a single image and associated sparse depth map, that closes the performance gap in a single pass. We first present a study on how the domain shift in each data modality affects model performance. Based on our observations that the sparse depth modality exhibits a much smaller covariate shift than the image, we design an embedding module trained in the source domain that preserves a mapping from features encoding only sparse depth to those encoding image and sparse depth. During test time, sparse depth features are projected using this map as a proxy for source domain features and are used as guidance to train a set of auxiliary parameters (i.e., adaptation layer) to align image and sparse depth features from the target test domain to that of the source domain. We evaluate our method on indoor and outdoor scenarios and show that it improves over baselines by an average of 21.1%.

The ability to associate touch with other modalities has huge implications for humans and computational systems. However, multimodal learning with touch remains challenging due to the expensive data collection process and non-standardized sensor outputs. We introduce UniTouch, a unified tactile model for vision-based touch sensors connected to multiple modalities, including vision, language, and sound. We achieve this by aligning our UniTouch embeddings to pretrained image embeddings already associated with a variety of other modalities. We further propose learnable sensor-specific tokens, allowing the model to learn from a set of heterogeneous tactile sensors, all at the same time. UniTouch is capable of conducting various touch sensing tasks in the zero-shot setting, from robot grasping prediction to touch image question answering. To the best of our knowledge, UniTouch is the first to demonstrate such capabilities. Project page: https://cfeng16.github.io/UniTouch/

We discover the presence of quantization artifacts in Vision Transformers (ViTs), which arise due to the image tokenization step inherent in these architectures. These artifacts result in coarsely quantized features, which negatively impact performance, especially on downstream dense prediction tasks. We present a zero-shot method to improve how pre-trained ViTs handle spatial quantization. In particular, we propose to ensemble the features obtained from perturbing input images via sub-token spatial translations, inspired by Stochastic Resonance, a method traditionally applied to climate dynamics and signal processing. We term our method ``Stochastic Resonance Transformer" (SRT), which we show can effectively super-resolve features of pre-trained ViTs, capturing more of the local fine-grained structures that might otherwise be neglected as a result of tokenization. SRT can be applied at any layer, on any task, and does not require any fine-tuning. The advantage of the former is evident when applied to monocular depth prediction, where we show that ensembling model outputs are detrimental while applying SRT on intermediate ViT features outperforms the baseline models by an average of 4.7% and 14.9% on the RMSE and RMSE-log metrics across three different architectures. When applied to semi-supervised video object segmentation, SRT also improves over the baseline models uniformly across all metrics, and by an average of 2.4% in F&J score. We further show that these quantization artifacts can be attenuated to some extent via self-distillation. On the unsupervised salient region segmentation, SRT improves upon the base model by an average of 2.1% on the maxF metric. Finally, despite operating purely on pixel-level features, SRT generalizes to non-dense prediction tasks such as image retrieval and object discovery, yielding consistent improvements of up to 2.6% and 1.0% respectively.

Depth perception models are typically trained on non-interactive datasets with predefined camera trajectories. However, this often introduces systematic biases into the learning process correlated to specific camera paths chosen during data acquisition. In this paper, we investigate the role of how data is collected for learning depth completion, from a robot navigation perspective, by leveraging 3D interactive environments. First, we evaluate four depth completion models trained on data collected using conventional navigation techniques. Our key insight is that existing exploration paradigms do not necessarily provide task-specific data points to achieve competent unsupervised depth completion learning. We then find that data collected with respect to photometric reconstruction has a direct positive influence on model performance. As a result, we develop an active, task-informed, depth uncertainty-based motion planning approach for learning depth completion, which we call DEpth Uncertainty-guided eXploration (DEUX). Training with data collected by our approach improves depth completion by an average greater than 18% across four depth completion models compared to existing exploration methods on the MP3D test set. We show that our approach further improves zero-shot generalization, while offering new insights into integrating robot learning-based depth estimation.

We study the effect of adversarial perturbations of images on deep stereo matching networks for the disparity estimation task. We present a method to craft a single set of perturbations that, when added to any stereo image pair in a dataset, can fool a stereo network to significantly alter the perceived scene geometry. Our perturbation images are "universal" in that they not only corrupt estimates of the network on the dataset they are optimized for, but also generalize to stereo networks with different architectures across different datasets. We evaluate our approach on multiple public benchmark datasets and show that our perturbations can increase D1-error (akin to fooling rate) of state-of-the-art stereo networks from 1% to as much as 87%. We investigate the effect of perturbations on the estimated scene geometry and identify object classes that are most vulnerable. Our analysis on the activations of registered points between left and right images led us to find that certain architectural components, i.e. deformable convolution and explicit matching, can increase robustness against adversaries. We demonstrate that by simply designing networks with such components, one can reduce the effect of adversaries by up to 60.5%, which rivals the robustness of networks fine-tuned with costly adversarial data augmentation.