NeurIPS 2024 论文列表

We consider the task of active geo-localization (AGL) in which an agent uses a sequence of visual cues observed during aerial navigation to find a target specified through multiple possible modalities. This could emulate a UAV involved in a search-and-rescue operation navigating through an area, observing a stream of aerial images as it goes. The AGL task is associated with two important challenges. Firstly, an agent must deal with a goal specification in one of multiple modalities (e.g., through a natural language description) while the search cues are provided in other modalities (aerial imagery). The second challenge is limited localization time (e.g., limited battery life, urgency) so that the goal must be localized as efficiently as possible, i.e. the agent must effectively leverage its sequentially observed aerial views when searching for the goal. To address these challenges, we propose GOMAA-Geo -- a goal modality agnostic active geo-localization agent -- for zero-shot generalization between different goal modalities. Our approach combines cross-modality contrastive learning to align representations across modalities with supervised foundation model pretraining and reinforcement learning to obtain highly effective navigation and localization policies. Through extensive evaluations, we show that GOMAA-Geo outperforms alternative learnable approaches and that it generalizes across datasets -- e.g., to disaster-hit areas without seeing a single disaster scenario during training -- and goal modalities -- e.g., to ground-level imagery or textual descriptions, despite only being trained with goals specified as aerial views. Our code is available at: https://github.com/mvrl/GOMAA-Geo.

Serialization-based methods, which serialize the 3D voxels and group them into multiple sequences before inputting to Transformers, have demonstrated their effectiveness in 3D object detection. However, serializing 3D voxels into 1D sequences will inevitably sacrifice the voxel spatial proximity. Such an issue is hard to be addressed by enlarging the group size with existing serialization-based methods due to the quadratic complexity of Transformers with feature sizes. Inspired by the recent advances of state space models (SSMs), we present a Voxel SSM, termed as Voxel Mamba, which employs a group-free strategy to serialize the whole space of voxels into a single sequence. The linear complexity of SSMs encourages our group-free design, alleviating the loss of spatial proximity of voxels. To further enhance the spatial proximity, we propose a Dual-scale SSM Block to establish a hierarchical structure, enabling a larger receptive field in the 1D serialization curve, as well as more complete local regions in 3D space. Moreover, we implicitly apply window partition under the group-free framework by positional encoding, which further enhances spatial proximity by encoding voxel positional information. Our experiments on Waymo Open Dataset and nuScenes dataset show that Voxel Mamba not only achieves higher accuracy than state-of-the-art methods, but also demonstrates significant advantages in computational efficiency. The source code is available at https://github.com/gwenzhang/Voxel-Mamba.

We consider the training of the first layer of vision models and notice the clear relationship between pixel values and gradient update magnitudes: the gradients arriving at the weights of a first layer are by definition directly proportional to (normalized) input pixel values. Thus, an image with low contrast has a smaller impact on learning than an image with higher contrast, and a very bright or very dark image has a stronger impact on the weights than an image with moderate brightness. In this work, we propose performing gradient descent on the embeddings produced by the first layer of the model. However, switching to discrete inputs with an embedding layer is not a reasonable option for vision models. Thus, we propose the conceptual procedure of (i) a gradient descent step on first layer activations to construct an activation proposal, and (ii) finding the optimal weights of the first layer, i.e., those weights which minimize the squared distance to the activation proposal. We provide a closed form solution of the procedure and adjust it for robust stochastic training while computing everything efficiently. Empirically, we find that TrAct (Training Activations) speeds up training by factors between 1.25x and 4x while requiring only a small computational overhead. We demonstrate the utility of TrAct with different optimizers for a range of different vision models including convolutional and transformer architectures.

Object tracking is a fundamental task in computer vision, requiring the localization of objects of interest across video frames. Diffusion models have shown remarkable capabilities in visual generation, making them well-suited for addressing several requirements of the tracking problem. This work proposes a novel diffusion-based methodology to formulate the tracking task. Firstly, their conditional process allows for injecting indications of the target object into the generation process. Secondly, diffusion mechanics can be developed to inherently model temporal correspondences, enabling the reconstruction of actual frames in video. However, existing diffusion models rely on extensive and unnecessary mapping to a Gaussian noise domain, which can be replaced by a more efficient and stable interpolation process. Our proposed interpolation mechanism draws inspiration from classic image-processing techniques, offering a more interpretable, stable, and faster approach tailored specifically for the object tracking task. By leveraging the strengths of diffusion models while circumventing their limitations, our Diffusion-based INterpolation TrackeR (DINTR) presents a promising new paradigm and achieves a superior multiplicity on seven benchmarks across five indicator representations.

3D Gaussian Splatting has shown fast and high-quality rendering results in static scenes by leveraging dense 3D prior and explicit representations. Unfortunately, the benefits of the prior and representation do not involve novel view synthesis for dynamic motions. Ironically, this is because the main barrier is the reliance on them, which requires increasing training and rendering times to account for dynamic motions. In this paper, we design Explicit 4D Gaussian Splatting (Ex4DGS). Our key idea is to firstly separate static and dynamic Gaussians during training, and to explicitly sample positions and rotations of the dynamic Gaussians at sparse timestamps. The sampled positions and rotations are then interpolated to represent both spatially and temporally continuous motions of objects in dynamic scenes as well as reducing computational cost. Additionally, we introduce a progressive training scheme and a point-backtracking technique that improves Ex4DGS's convergence. We initially train Ex4DGS using short timestamps and progressively extend timestamps, which makes it work well with a few point clouds. The point-backtracking is used to quantify the cumulative error of each Gaussian over time, enabling the detection and removal of erroneous Gaussians in dynamic scenes. Comprehensive experiments on various scenes demonstrate the state-of-the-art rendering quality from our method, achieving fast rendering of 62 fps on a single 2080Ti GPU.

This paper investigates the 3D domain generalization (3DDG) ability of large 3D models based on prevalent prompt learning. Recent works demonstrate the performances of 3D point cloud recognition can be boosted remarkably by parameter-efficient prompt tuning. However, we observe that the improvement on downstream tasks comes at the expense of a severe drop in 3D domain generalization. To resolve this challenge, we present a comprehensive regulation framework that allows the learnable prompts to actively interact with the well-learned general knowledge in large 3D models to maintain good generalization. Specifically, the proposed framework imposes multiple explicit constraints on the prompt learning trajectory by maximizing the mutual agreement between task-specific predictions and task-agnostic knowledge. We design the regulation framework as a plug-and-play module to embed into existing representative large 3D models. Surprisingly, our method not only realizes consistently increasing generalization ability but also enhances task-specific 3D recognition performances across various 3DDG benchmarks by a clear margin. Considering the lack of study and evaluation on 3DDG, we also create three new benchmarks, namely base-to-new, cross-dataset and few-shot generalization benchmarks, to enrich the field and inspire future research. Code and benchmarks are available at \url{https://github.com/auniquesun/Point-PRC}.

Large Multi-modal Models (LMMs) have recently demonstrated remarkable abilities in visual context understanding and coherent response generation. However, alongside these advancements, the issue of hallucinations has emerged as a significant challenge, producing erroneous responses that are unrelated to the visual contents. In this paper, we introduce a novel contrastive-based decoding method, COuntering DEscription Contrastive Decoding (CODE), which leverages self-generated descriptions as contrasting references during the decoding phase of LMMs to address hallucination issues. CODE utilizes the comprehensive descriptions from model itself as visual counterpart to correct and improve response alignment with actual visual content. By dynamically adjusting the information flow and distribution of next-token predictions in the LMM's vocabulary, CODE enhances the coherence and informativeness of generated responses. Extensive experiments demonstrate that our method significantly reduces hallucinations and improves cross-modal consistency across various benchmarks and cutting-edge LMMs. Our method provides a simple yet effective decoding strategy that can be integrated to existing LMM frameworks without additional training.

Audio deepfake detection (ADD) is crucial to combat the misuse of speech synthesized by generative AI models. Existing ADD models suffer from generalization issues to unseen attacks, with a large performance discrepancy between in-domain and out-of-domain data. Moreover, the black-box nature of existing models limits their use in real-world scenarios, where explanations are required for model decisions. To alleviate these issues, we introduce a new ADD model that explicitly uses the Style-LInguistics Mismatch (SLIM) in fake speech to separate them from real speech. SLIM first employs self-supervised pretraining on only real samples to learn the style-linguistics dependency in the real class. The learned features are then used in complement with standard pretrained acoustic features (e.g., Wav2vec) to learn a classifier on the real and fake classes. When the feature encoders are frozen, SLIM outperforms benchmark methods on out-of-domain datasets while achieving competitive results on in-domain data. The features learned by SLIM allow us to quantify the (mis)match between style and linguistic content in a sample, hence facilitating an explanation of the model decision.

Tiny machine learning (TinyML) aims to run ML models on small devices and is increasingly favored for its enhanced privacy, reduced latency, and low cost. Recently, the advent of tiny AI accelerators has revolutionized the TinyML field by significantly enhancing hardware processing power. These accelerators, equipped with multiple parallel processors and dedicated per-processor memory instances, offer substantial performance improvements over traditional microcontroller units (MCUs). However, their limited data memory often necessitates downsampling input images, resulting in accuracy degradation. To address this challenge, we propose Data channel EXtension (DEX), a novel approach for efficient CNN execution on tiny AI accelerators. DEX incorporates additional spatial information from original images into input images through patch-wise even sampling and channel-wise stacking, effectively extending data across input channels. By leveraging underutilized processors and data memory for channel extension, DEX facilitates parallel execution without increasing inference latency. Our evaluation with four models and four datasets on tiny AI accelerators demonstrates that this simple idea improves accuracy on average by 3.5%p while keeping the inference latency the same on the AI accelerator. The source code is available at https://github.com/Nokia-Bell-Labs/data-channel-extension.

In this paper, we tackle the challenge of generating high-quality hash codes for cross-modal retrieval in the presence of incomplete labels, which creates uncertainty in distinguishing between positive and negative pairs. Vision-language models such as CLIP offer a potential solution by providing generic knowledge for missing label recovery, yet their zero-shot performance remains insufficient. To address this, we propose a novel Prompt Contrastive Recovery approach, \textbf{PCRIL}, which progressively identifies promising positive classes from unknown label sets and recursively searches for other relevant labels. Identifying unknowns is nontrivial due to the fixed and long-tailed patterns of positive label sets in training data, which hampers the discovery of new label combinations. Therefore, we consider each subset of positive labels and construct three types of negative prompts through deletion, addition, and replacement for prompt learning. The augmented supervision guides the model to measure the completeness of label sets, thus facilitating the subsequent greedy tree search for label completion. We also address extreme cases of significant unknown labels and lack of negative pairwise supervision by deriving two augmentation strategies: seeking unknown-complementary samples for mixup and random flipping for negative labels. Extensive experiments reveal the vulnerability of current methods and demonstrate the effectiveness of PCRIL, achieving an average 12\% mAP improvement to the current SOTA across all datasets. Our code is available at https://github.com/E-Galois/PCRIL.

We introduce Referring Human Pose and Mask Estimation (R-HPM) in the wild, where either a text or positional prompt specifies the person of interest in an image. This new task holds significant potential for human-centric applications such as assistive robotics and sports analysis. In contrast to previous works, R-HPM (i) ensures high-quality, identity-aware results corresponding to the referred person, and (ii) simultaneously predicts human pose and mask for a comprehensive representation. To achieve this, we introduce a large-scale dataset named RefHuman, which substantially extends the MS COCO dataset with additional text and positional prompt annotations. RefHuman includes over 50,000 annotated instances in the wild, each equipped with keypoint, mask, and prompt annotations. To enable prompt-conditioned estimation, we propose the first end-to-end promptable approach named UniPHD for R-HPM. UniPHD extracts multimodal representations and employs a proposed pose-centric hierarchical decoder to process (text or positional) instance queries and keypoint queries, producing results specific to the referred person. Extensive experiments demonstrate that UniPHD produces quality results based on user-friendly prompts and achieves top-tier performance on RefHuman val and MS COCO val2017.

Most large multimodal models (LMMs) are implemented by feeding visual tokens as a sequence into the first layer of a large language model (LLM). The resulting architecture is simple but significantly increases computation and memory costs, as it has to handle a large number of additional tokens in its input layer. This paper presents a new architecture *DeepStack* for LMMs. Considering $N$ layers in the language and vision transformer of LMMs, we stack the visual tokens into $N$ groups and feed each group to its aligned transformer layer from bottom to top. Surprisingly, this simple method greatly enhances the power of LMMs to model interactions among visual tokens across layers but with minimal additional cost. We apply *DeepStack* to both language and vision transformer in LMMs, and validate the effectiveness of *DeepStack* LMMs with extensive empirical results. Using the same context length, our DeepStack 7B and 13B parameters surpass their counterparts by 2.7 and 2.9 on average across 9 benchmarks, respectively. Using only one-fifth of the context length, DeepStack rivals closely to the counterparts that use the full context length. These gains are particularly pronounced on high-resolution tasks, *e.g.*, 4.2, 11.0, and 4.0 improvements on TextVQA, DocVQA, and InfoVQA compared to LLaVA-1.5-7B, respectively. We further apply *DeepStack* to vision transformer layers, which brings us a similar amount of improvements, 3.8 on average compared with LLaVA-1.5-7B.

Generative text-to-image models enable us to synthesize unlimited amounts of images in a controllable manner, spurring many recent efforts to train vision models with synthetic data. However, every synthetic image ultimately originates from the upstream data used to train the generator. Does the intermediate generator provide additional information over directly training on relevant parts of the upstream data? Grounding this question in the setting of image classification, we compare finetuning on task-relevant, targeted synthetic data generated by Stable Diffusion---a generative model trained on the LAION-2B dataset---against finetuning on targeted real images retrieved directly from LAION-2B. We show that while synthetic data can benefit some downstream tasks, it is universally matched or outperformed by real data from the simple retrieval baseline. Our analysis suggests that this underperformance is partially due to generator artifacts and inaccurate task-relevant visual details in the synthetic images. Overall, we argue that targeted retrieval is a critical baseline to consider when training with synthetic data---a baseline that current methods do not yet surpass. We release code, data, and models at [https://github.com/scottgeng00/unmet-promise/](https://github.com/scottgeng00/unmet-promise).

Video Anomaly Detection (VAD) finds widespread applications in security surveillance, traffic monitoring, industrial monitoring, and healthcare. Despite extensive research efforts, there remains a lack of concise reviews that provide insightful guidance for researchers. Such reviews would serve as quick references to grasp current challenges, research trends, and future directions. In this paper, we present such a review, examining models and datasets from various perspectives. We emphasize the critical relationship between model and dataset, where the quality and diversity of datasets profoundly influence model performance, and dataset development adapts to the evolving needs of emerging approaches. Our review identifies practical issues, including the absence of comprehensive datasets with diverse scenarios. To address this, we introduce a new dataset, Multi-Scenario Anomaly Detection (MSAD), comprising 14 distinct scenarios captured from various camera views. Our dataset has diverse motion patterns and challenging variations, such as different lighting and weather conditions, providing a robust foundation for training superior models. We conduct an in-depth analysis of recent representative models using MSAD and highlight its potential in addressing the challenges of detecting anomalies across diverse and evolving surveillance scenarios.

3D Gaussian splatting (3DGS) has shown promising results in image rendering and surface reconstruction. However, its potential in volumetric reconstruction tasks, such as X-ray computed tomography, remains under-explored. This paper introduces R$^2$-Gaussian, the first 3DGS-based framework for sparse-view tomographic reconstruction. By carefully deriving X-ray rasterization functions, we discover a previously unknown \emph{integration bias} in the standard 3DGS formulation, which hampers accurate volume retrieval. To address this issue, we propose a novel rectification technique via refactoring the projection from 3D to 2D Gaussians. Our new method presents three key innovations: (1) introducing tailored Gaussian kernels, (2) extending rasterization to X-ray imaging, and (3) developing a CUDA-based differentiable voxelizer. Experiments on synthetic and real-world datasets demonstrate that our method outperforms state-of-the-art approaches in accuracy and efficiency. Crucially, it delivers high-quality results in 4 minutes, which is 12$\times$ faster than NeRF-based methods and on par with traditional algorithms.

This paper presents MaVEn, an innovative Multi-granularity Visual Encoding framework designed to enhance the capabilities of Multimodal Large Language Models (MLLMs) in multi-image reasoning. Current MLLMs primarily focus on single-image visual understanding, limiting their ability to interpret and integrate information across multiple images. MaVEn addresses this limitation by combining discrete visual symbol sequences, which abstract coarse-grained semantic concepts, with traditional continuous representation sequences that model fine-grained features. This dual approach bridges the semantic gap between visual and textual data, thereby improving the model's ability to process and interpret information from multiple images effectively. Additionally, we design a dynamic reduction mechanism by for long-sequence continuous features to enhance multi-image processing efficiency. Experimental results demonstrate that MaVEn significantly enhances MLLMs' understanding in complex multi-image scenarios, while also improving performance in single-image contexts.

Existing state-of-the-art dense object detection techniques tend to produce a large number of false positive detections on difficult images with complex scenes because they focus on ensuring a high recall. To improve the detection accuracy, we propose an Adaptive Important Region Selection (AIRS) framework guided by Evidential Q-learning coupled with a uniquely designed reward function. Inspired by human visual attention, our detection model conducts object search in a top-down, hierarchical fashion. It starts from the top of the hierarchy with the coarsest granularity and then identifies the potential patches likely to contain objects of interest. It then discards non-informative patches and progressively moves downward on the selected ones for a fine-grained search. The proposed evidential Q-learning systematically encodes epistemic uncertainty in its evidential-Q value to encourage the exploration of unknown patches, especially in the early phase of model training. In this way, the proposed model dynamically balances exploration-exploitation to cover both highly valuable and informative patches. Theoretical analysis and extensive experiments on multiple datasets demonstrate that our proposed framework outperforms the SOTA models.

Recently, Gaussian splatting has received more and more attention in the field of static scene rendering. Due to the low computational overhead and inherent flexibility of explicit representations, plane-based explicit methods are popular ways to predict deformations for Gaussian-based dynamic scene rendering models. However, plane-based methods rely on the inappropriate low-rank assumption and excessively decompose the space-time 4D encoding, resulting in overmuch feature overlap and unsatisfactory rendering quality. To tackle these problems, we propose Grid4D, a dynamic scene rendering model based on Gaussian splatting and employing a novel explicit encoding method for the 4D input through the hash encoding. Different from plane-based explicit representations, we decompose the 4D encoding into one spatial and three temporal 3D hash encodings without the low-rank assumption. Additionally, we design a novel attention module that generates the attention scores in a directional range to aggregate the spatial and temporal features. The directional attention enables Grid4D to more accurately fit the diverse deformations across distinct scene components based on the spatial encoded features. Moreover, to mitigate the inherent lack of smoothness in explicit representation methods, we introduce a smooth regularization term that keeps our model from the chaos of deformation prediction. Our experiments demonstrate that Grid4D significantly outperforms the state-of-the-art models in visual quality and rendering speed.

Large vision-language models (LVLMs) have recently achieved rapid progress, sparking numerous studies to evaluate their multi-modal capabilities. However, we dig into current evaluation works and identify two primary issues: 1) Visual content is unnecessary for many samples. The answers can be directly inferred from the questions and options, or the world knowledge embedded in LLMs. This phenomenon is prevalent across current benchmarks. For instance, GeminiPro achieves 42.7% on the MMMU benchmark without any visual input, and outperforms the random choice baseline across six benchmarks near 24% on average. 2) Unintentional data leakage exists in LLM and LVLM training. LLM and LVLM could still answer some visual-necessary questions without visual content, indicating the memorizing of these samples within large-scale training data. For example, Sphinx-X-MoE gets 43.6% on MMMU without accessing images, surpassing its LLM backbone with 17.9%. Both problems lead to misjudgments of actual multi-modal gains and potentially misguide the study of LVLM. To this end, we present MMStar, an elite vision-indispensable multi-modal benchmark comprising 1,500 samples meticulously selected by humans. MMStar benchmarks 6 core capabilities and 18 detailed axes, aiming to evaluate LVLMs' multi-modal capacities with carefully balanced and purified samples. These samples are first roughly selected from current benchmarks with an automated pipeline, human review is then involved to ensure each curated sample exhibits visual dependency, minimal data leakage, and requires advanced multi-modal capabilities. Moreover, two metrics are developed to measure data leakage and actual performance gain in multi-modal training. We evaluate 16 leading LVLMs on MMStar to assess their multi-modal capabilities, and on 7 benchmarks with the proposed metrics to investigate their data leakage and actual multi-modal gain.