OpEnCam: Optical Encryption Camera

Lensless cameras multiplex the incoming light before it is recorded by the sensor. This ability to multiplex the incoming light has led to the development of ultra-thin, high-speed, and single-shot 3D imagers. Recently, there have been various attempts at demonstrating another useful aspect of lensless cameras – their ability to preserve the privacy of a scene by capturing encrypted measurements. However, existing lensless camera designs suffer numerous inherent privacy vulnerabilities. To demonstrate this, we develop the first comprehensive attack model for encryption cameras, and propose OpEnCam — a novel lensless OPtical ENcryption CAmera design that overcomes these vulnerabilities. OpEnCam encrypts the incoming light before capturing it using the modulating ability of optical masks. Recovery of the original scene from an OpEnCam measurement is possible only if one has access to the camera’s encryption key, defined by the unique optical elements of each camera. Our OpEnCam design introduces two major improvements over existing lensless camera designs – (a) the use of two co-axially located optical masks, one stuck to the sensor and the other a few millimeters above the sensor and (b) the design of mask patterns, which are derived heuristically from signal processing ideas. We show, through experiments, that OpEnCam is robust against a range of attack types while still maintaining the imaging capabilities of existing lensless cameras. We validate the efficacy of OpEnCam using simulated and real data. Finally, we built and tested a prototype in the lab for proof-of-concept.

LLM-ASSIST: Enhancing Closed-Loop Planning with Language-Based Reasoning

Although planning is a crucial component of the autonomous driving stack, researchers have yet to develop robust planning algorithms that are capable of safely handling the diverse range of possible driving scenarios. Learning-based planners suffer from overfitting and poor long-tail performance. On the other hand, rule-based planners generalize well, but might fail to handle scenarios that require complex driving maneuvers. To address these limitations, we investigate the possibility of leveraging the common-sense reasoning capabilities of Large Language Models (LLMs) such as GPT4 and Llama2 to generate plans for self-driving vehicles. In particular, we develop a novel hybrid planner that leverages a conventional rule-based planner in conjunction with an LLM-based planner. Guided by commonsense reasoning abilities of LLMs, our approach navigates complex scenarios which existing planners struggle with, produces well-reasoned outputs while also remaining grounded through working alongside the rule-based approach. Through extensive evaluation on the nuPlan benchmark, we achieve state-of-the-art performance, outperforming all existing pure learning- and rule-based methods across most metrics. Our code will be available at

Controllable Safety-Critical Closed-loop Traffic Simulation via Guided Diffusion

Evaluating the performance of autonomous vehicle planning algorithms necessitates simulating long-tail traffic scenarios. Traditional methods for generating safety-critical scenarios often fall short in realism and controllability. Furthermore, these techniques generally neglect the dynamics of agent interactions. To mitigate these limitations, we introduce a novel closed-loop simulation framework rooted in guided diffusion models. Our approach yields two distinct advantages: 1) the generation of realistic long-tail scenarios that closely emulate real-world conditions, and 2) enhanced controllability, enabling more comprehensive and interactive evaluations. We achieve this through novel guidance objectives that enhance road progress while lowering collision and off-road rates. We develop a novel approach to simulate safety-critical scenarios through an adversarial term in the denoising process, which allows the adversarial agent to challenge a planner with plausible maneuvers, while all agents in the scene exhibit reactive and realistic behaviors. We validate our framework empirically using the NuScenes dataset, demonstrating improvements in both realism and controllability. These findings affirm that guided diffusion models provide a robust and versatile foundation for safety-critical, interactive traffic simulation, extending their utility across the broader landscape of autonomous driving. For additional resources and demonstrations, visit our project page at

Domain Generalization Guided by Gradient Signal to Noise Ratio of Parameters

Overfitting to the source domain is a common issue in gradient-based training of deep neural networks. To compensate for the over-parameterized models, numerous regularization techniques have been introduced such as those based on dropout. While these methods achieve significant improvements on classical benchmarks such as ImageNet, their performance diminishes with the introduction of domain shift in the test set i.e. when the unseen data comes from a significantly different distribution. In this paper, we move away from the classical approach of Bernoulli sampled dropout mask construction and propose to base the selection on gradient-signal-to-noise ratio (GSNR) of network’s parameters. Specifically, at each training step, parameters with high GSNR will be discarded. Furthermore, we alleviate the burden of manually searching for the optimal dropout ratio by leveraging a meta-learning approach. We evaluate our method on standard domain generalization benchmarks and achieve competitive results on classification and face anti-spoofing problems.

Efficient Controllable Multi-Task Architectures

Efficient Controllable Multi-Task Architectures We aim to train a multi-task model such that users can adjust the desired compute budget and relative importance of task performances after deployment, without retraining. This enables optimizing performance for dynamically varying user needs, without heavy computational overhead to train and save models for various scenarios. To this end, we propose a multi-task model consisting of a shared encoder and task-specific decoders where both encoder and decoder channel widths are slimmable. Our key idea is to control the task importance by varying the capacities of task-specific decoders, while controlling the total computational cost by jointly adjusting the encoder capacity. This improves overall accuracy by allowing a stronger encoder for a given budget, increases control over computational cost, and delivers high-quality slimmed sub-architectures based on user’s constraints. Our training strategy involves a novel `Configuration-Invariant Knowledge Distillation’ loss that enforces backbone representations to be invariant under different runtime width configurations to enhance accuracy. Further, we present a simple but effective search algorithm that translates user constraints to runtime width configurations of both the shared encoder and task decoders, for sampling the sub-architectures. The key rule for the search algorithm is to provide a larger computational budget to the higher preferred task decoder, while searching a shared encoder configuration that enhances the overall MTL performance. Various experiments on three multi-task benchmarks (PASCALContext, NYUDv2, and CIFAR100-MTL) with diverse backbone architectures demonstrate the advantage of our approach. For example, our method shows a higher controllability by 33.5% in the NYUD-v2 dataset over prior methods, while incurring much less compute cost.

Improving Pseudo Labels for Open-Vocabulary Object Detection

Improving Pseudo Labels for Open-Vocabulary Object Detection Recent studies show promising performance in open-vocabulary object detection (OVD) using pseudo labels (PLs) from pretrained vision and language models (VLMs). However, PLs generated by VLMs are extremely noisy due to the gap between the pretraining objective of VLMs and OVD, which blocks further advances on PLs. In this paper, we aim to reduce the noise in PLs and propose a method called online Self-training And a Split-and-fusion head for OVD (SAS-Det). First, the self-training finetunes VLMs to generate high quality PLs while prevents forgetting the knowledge learned in the pretraining. Second, a split-and-fusion (SAF) head is designed to remove the noise in localization of PLs, which is usually ignored in existing methods. It also fuses complementary knowledge learned from both precise ground truth and noisy pseudo labels to boost the performance. Extensive experiments demonstrate SAS-Det is both efficient and effective. Our pseudo labeling is 3 times faster than prior methods. SAS-Det outperforms prior state-of-the-art models of the same scale by a clear margin and achieves 37.4 AP50 and 27.3 APr on novel categories of the COCO and LVIS benchmarks, respectively.

Q: How to Specialize Large Vision-Language Models to Data-Scarce VQA Tasks? A: Self-Train on Unlabeled Images!

Q: How to Specialize Large Vision-Language Models to Data-Scarce VQA Tasks? A: Self-Train on Unlabeled Images! Finetuning a large vision language model (VLM) on a target dataset after large scale pretraining is a dominant paradigm in visual question answering (VQA). Datasets for specialized tasks such as knowledge-based VQA or VQA in non natural-image domains are orders of magnitude smaller than those for general-purpose VQA. While collecting additional labels for specialized tasks or domains can be challenging, unlabeled images are often available. We introduce SelTDA (Self-Taught Data Augmentation), a strategy for finetuning large VLMs on small-scale VQA datasets. SelTDA uses the VLM and target dataset to build a teacher model that can generate question-answer pseudolabels directly conditioned on an image alone, allowing us to pseudolabel unlabeled images. SelTDA then finetunes the initial VLM on the original dataset augmented with freshly pseudolabeled images. We describe a series of experiments showing that our self-taught data augmentation increases robustness to adversarially searched questions, counterfactual examples, and rephrasings, it improves domain generalization, and results in greater retention of numerical reasoning skills. The proposed strategy requires no additional annotations or architectural modifications, and is compatible with any modern encoder-decoder multimodal transformer. Code available at

NeurOCS: Neural NOCS Supervision for Monocular 3D Object Localization

NeurOCS: Neural NOCS Supervision for Monocular 3D Object Localization Monocular 3D object localization in driving scenes is a crucial task, but challenging due to its ill-posed nature. Estimating 3D coordinates for each pixel on the object surface holds great potential as it provides dense 2D-3D geometric constraints for the underlying PnP problem. However, high-quality ground truth supervision is not available in driving scenes due to sparsity and various artifacts of Lidar data, as well as the practical infeasibility of collecting per-instance CAD models. In this work, we present NeurOCS, a framework that uses instance masks and 3D boxes as input to learn 3D object shapes by means of differentiable rendering, which further serves as supervision for learning dense object coordinates. Our approach rests on insights in learning a category-level shape prior directly from real driving scenes, while properly handling single-view ambiguities. Furthermore, we study and make critical design choices to learn object coordinates more effectively from an object-centric view. Altogether, our framework leads to new state-of-the-art in monocular 3D localization that ranks 1st on the KITTI-Object benchmark among published monocular methods.

Split to Learn: Gradient Split for Multi-Task Human Image Analysis

Split to Learn: Gradient Split for Multi-Task Human Image Analysis This paper presents an approach to train a unified deep network that simultaneously solves multiple human-related tasks. A multi-task framework is favorable for sharing information across tasks under restricted computational resources. However, tasks not only share information but may also compete for resources and conflict with each other, making the optimization of shared parameters difficult and leading to suboptimal performance. We propose a simple but effective training scheme called GradSplit that alleviates this issue by utilizing asymmetric inter-task relations. Specifically, at each convolution module, it splits features into T groups for T tasks and trains each group only using the gradient back-propagated from the task losses with which it does not have conflicts. During training, we apply GradSplit to a series of convolution modules. As a result, each module is trained to generate a set of task-specific features using the shared features from the previous module. This enables a network to use complementary information across tasks while circumventing gradient conflicts. Experimental results show that GradSplit achieves a better accuracy-efficiency trade-off than existing methods. It minimizes accuracy drop caused by task conflicts while significantly saving compute resources in terms of both FLOPs and memory at inference. We further show that GradSplit achieves higher cross-dataset accuracy compared to single-task and other multi-task networks.

Single-Stream Multi-level Alignment for Vision-Language Pretraining

Single-Stream Multi-level Alignment for Vision-Language Pretraining Self-supervised vision-language pretraining from pure images and text with a contrastive loss is effective, but ignores fine-grained alignment due to a dual-stream architecture that aligns image and text representations only on a global level. Earlier, supervised, non-contrastive methods were capable of finer-grained alignment, but required dense annotations that were not scalable. We propose a single stream architecture that aligns images and language at multiple levels: global, fine-grained patch-token, and conceptual/semantic, using two novel tasks: symmetric cross-modality reconstruction (XMM) and a pseudo-labeled key word prediction (PSL). In XMM, we mask input tokens from one modality and use cross-modal information to reconstruct the masked token, thus improving fine-grained alignment between the two modalities. In PSL, we use attention to select keywords in a caption, use a momentum encoder to recommend other important keywords that are missing from the caption but represented in the image, and then train the visual encoder to predict the presence of those keywords, helping it learn semantic concepts that are essential for grounding a textual token to an image region. We demonstrate competitive performance and improved data efficiency on image-text retrieval, grounding, visual question answering/reasoning against larger models and models trained on more data. Code and models available at