Action recognition models have achieved impressive results by incorporating scene-level annotations, such as objects, their relations, 3D structure, and more. However, obtaining annotations of scene structure for videos requires a significant amount of effort to gather and annotate, making these methods expensive to train. In contrast, synthetic datasets generated by graphics engines provide powerful alternatives for generating scene-level annotations across multiple tasks. In this work, we propose an approach to leverage synthetic scene data for improving video understanding. We present a multi-task prompt learning approach for video transformers, where a shared video transformer backbone is enhanced by a small set of specialized parameters for each task. Specifically, we add a set of ``task prompts'', each corresponding to a different task, and let each prompt predict task-related annotations. This design allows the model to capture information shared among synthetic scene tasks as well as information shared between synthetic scene tasks and a real video downstream task throughout the entire network. We refer to this approach as ``Promptonomy'', since the prompts model a task-related structure. We propose the PromptonomyViT model (PViT), a video transformer that incorporates various types of scene-level information from synthetic data using the ``Promptonomy'' approach. PViT shows strong performance improvements on multiple video understanding tasks and datasets.

该技术报告描述了无回报（PNR）时间定位挑战的EGO4D点的SVIT方法。我们提出了一个学习框架的结构（简称SVIT），该结构证明了仅在训练过程中仅可用的少量图像的结构才能改善视频模型。SVIT依靠两个关键见解。首先，由于图像和视频都包含结构化信息，因此我们用一组\ emph {对象令牌}丰富了一个可以在图像和视频中使用的\ emph {对象令牌}的模型。其次，视频中各个帧的场景表示应与静止图像的场景表示“对齐”。这是通过“框架夹一致性”损失实现的，该损失可确保图像和视频之间结构化信息的流动。SVIT在挑战测试集上获得了强劲的性能，并具有0.656绝对时间定位误差。

最近的动作识别模型通过整合对象，其位置和互动来取得令人印象深刻的结果。但是，为每个框架获得密集的结构化注释是乏味且耗时的，使这些方法的训练昂贵且可扩展性较低。同时，如果可以在感兴趣的域内或之外使用一小部分带注释的图像，我们如何将它们用于下游任务的视频？我们提出了一个学习框架的结构（简称SVIT），该结构证明了仅在训练过程中仅可用的少量图像的结构才能改善视频模型。 SVIT依靠两个关键见解。首先，由于图像和视频都包含结构化信息，因此我们用一组\ emph {对象令牌}丰富了一个可以在图像和视频中使用的\ emph {对象令牌}的模型。其次，视频中各个帧的场景表示应与静止图像的场景表示“对齐”。这是通过\ emph {frame-clip一致性}损失来实现的，该损失可确保图像和视频之间结构化信息的流动。我们探索场景结构的特定实例化，即\ emph {手对象图}，由手和对象组成，其位置为节点，以及触点/no-contact的物理关系作为边缘。 SVIT在多个视频理解任务和数据集上显示出强烈的性能改进；它在EGO4D CVPR'22对象状态本地化挑战中赢得了第一名。对于代码和预算模型，请访问\ url {https://eladb3.github.io/svit/}的项目页面

最近，视频变压器在视频理解方面取得了巨大成功，超过了CNN性能;然而，现有的视频变换器模型不会明确地模拟对象，尽管对象对于识别操作至关重要。在这项工作中，我们呈现对象区域视频变换器（Orvit），一个\ emph {对象为中心}方法，它与直接包含对象表示的块扩展视频变压器图层。关键的想法是从早期层开始融合以对象形式的表示，并将它们传播到变压器层中，从而影响整个网络的时空表示。我们的orvit块由两个对象级流组成：外观和动态。在外观流中，“对象区域关注”模块在修补程序上应用自我关注和\ emph {对象区域}。以这种方式，Visual对象区域与统一修补程序令牌交互，并通过上下文化对象信息来丰富它们。我们通过单独的“对象 - 动态模块”进一步模型对象动态，捕获轨迹交互，并显示如何集成两个流。我们在四个任务和五个数据集中评估我们的模型：在某事物中的某些问题和几次射击动作识别，以及在AVA上的某些时空动作检测，以及在某种东西上的标准动作识别 - 某种东西 - 东西，潜水48和EPIC-Kitchen100。我们在考虑的所有任务和数据集中展示了强大的性能改进，展示了将对象表示的模型的值集成到变压器体系结构中。对于代码和预用模型，请访问项目页面\ url {https://roeiherz.github.io/orvit/}

Neural Representations have recently been shown to effectively reconstruct a wide range of signals from 3D meshes and shapes to images and videos. We show that, when adapted correctly, neural representations can be used to directly represent the weights of a pre-trained convolutional neural network, resulting in a Neural Representation for Neural Networks (NeRN). Inspired by coordinate inputs of previous neural representation methods, we assign a coordinate to each convolutional kernel in our network based on its position in the architecture, and optimize a predictor network to map coordinates to their corresponding weights. Similarly to the spatial smoothness of visual scenes, we show that incorporating a smoothness constraint over the original network's weights aids NeRN towards a better reconstruction. In addition, since slight perturbations in pre-trained model weights can result in a considerable accuracy loss, we employ techniques from the field of knowledge distillation to stabilize the learning process. We demonstrate the effectiveness of NeRN in reconstructing widely used architectures on CIFAR-10, CIFAR-100, and ImageNet. Finally, we present two applications using NeRN, demonstrating the capabilities of the learned representations.

In this short paper, we present our ongoing work on the veriFIRE project -- a collaboration between industry and academia, aimed at using verification for increasing the reliability of a real-world, safety-critical system. The system we target is an airborne platform for wildfire detection, which incorporates two deep neural networks. We describe the system and its properties of interest, and discuss our attempts to verify the system's consistency, i.e., its ability to continue and correctly classify a given input, even if the wildfire it describes increases in intensity. We regard this work as a step towards the incorporation of academic-oriented verification tools into real-world systems of interest.

Light is a complex-valued field. The intensity and phase of the field are affected by imaged objects. However, imaging sensors measure only real-valued non-negative intensities. This results in a nonlinear relation between the measurements and the unknown imaged objects. Moreover, the sensor readouts are corrupted by Poissonian-distributed photon noise. In this work, we seek the most probable object (or clear image), given noisy measurements, that is, maximizing the a-posteriori probability of the sought variables. Hence, we generalize annealed Langevin dynamics, tackling fundamental challenges in optical imaging, including phase recovery and Poisson (photon) denoising. We leverage deep neural networks, not for explicit recovery of the imaged object, but as an approximate gradient for a prior term. We show results on empirical data, acquired by a real experiment. We further show results of simulations.

Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now, massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources that are only available to well-resourced teams. In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets, ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.

Estimating uncertainty in image-to-image networks is an important task, particularly as such networks are being increasingly deployed in the biological and medical imaging realms. In this paper, we introduce a new approach to this problem based on masking. Given an existing image-to-image network, our approach computes a mask such that the distance between the masked reconstructed image and the masked true image is guaranteed to be less than a specified threshold, with high probability. The mask thus identifies the more certain regions of the reconstructed image. Our approach is agnostic to the underlying image-to-image network, and only requires triples of the input (degraded), reconstructed and true images for training. Furthermore, our method is agnostic to the distance metric used. As a result, one can use $L_p$-style distances or perceptual distances like LPIPS, which contrasts with interval-based approaches to uncertainty. Our theoretical guarantees derive from a conformal calibration procedure. We evaluate our mask-based approach to uncertainty on image colorization, image completion, and super-resolution tasks, demonstrating high quality performance on each.

Controllable image captioning models generate human-like image descriptions, enabling some kind of control over the generated captions. This paper focuses on controlling the caption length, i.e. a short and concise description or a long and detailed one. Since existing image captioning datasets contain mostly short captions, generating long captions is challenging. To address the shortage of long training examples, we propose to enrich the dataset with varying-length self-generated captions. These, however, might be of varying quality and are thus unsuitable for conventional training. We introduce a novel training strategy that selects the data points to be used at different times during the training. Our method dramatically improves the length-control abilities, while exhibiting SoTA performance in terms of caption quality. Our approach is general and is shown to be applicable also to paragraph generation.