Growing customer demand for smart solutions in robotics and augmented reality has attracted considerable attention to 3D object detection from point clouds. Yet, existing indoor datasets taken individually are too small and insufficiently diverse to train a powerful and general 3D object detection model. In the meantime, more general approaches utilizing foundation models are still inferior in quality to those based on supervised training for a specific task. In this work, we propose UniDet3D, a simple yet effective 3D object detection model, which is trained on a mixture of indoor datasets and is capable to work in various indoor environments. By unifying different label spaces, UniDet3D enables learning a strong representation across multiple datasets through a supervised joint training scheme. The proposed network architecture is built upon a vanilla transformer encoder, making it easy to run, customize and extend the prediction pipeline for practical use. Extensive experiments demonstrate that UniDet3D obtains significant gains over existing 3D object detection methods in 6 indoor benchmarks: ScanNet (+1.1 mAP50), S3DIS (+9.1 mAP50), ARKitScenes (+19.4 mAP25), MultiScan (+14.3 mAP50), 3RScan (+3.2 mAP50), and ScanNet++ (+2.7 mAP50).
论文检索
输入标题、作者或关键词，从 11,272 篇学术成果中精准定位
3D super-resolution aims to reconstruct high-fidelity 3D models from low-resolution (LR) multi-view images. Early studies primarily focused on single-image super-resolution (SISR) models to upsample LR images into high-resolution images. However, these methods often lack view consistency because they operate independently on each image. Although various post-processing techniques have been extensively explored to mitigate these inconsistencies, they have yet to fully resolve the issues. In this paper, we perform a comprehensive study of 3D super-resolution by leveraging video super-resolution (VSR) models. By utilizing VSR models, we ensure a higher degree of spatial consistency and can reference surrounding spatial information, leading to more accurate and detailed reconstructions. Our findings reveal that VSR models can perform remarkably well even on sequences that lack precise spatial alignment. Given this observation, we propose a simple yet practical approach to align LR images without involving fine-tuning or generating `smooth' trajectory from the trained 3D models over LR images. The experimental results show that the surprisingly simple algorithms can achieve the state-of-the-art results of 3D super-resolution tasks on standard benchmark datasets, such as the NeRF-synthetic and Mip-NeRF 360 datasets.
Geographic information is essential for modeling tasks in fields ranging from ecology to epidemiology. However, extracting relevant location characteristics for a given task can be challenging, often requiring expensive data fusion or distillation from massive global imagery datasets. To address this challenge, we introduce Satellite Contrastive Location-Image Pretraining (SatCLIP). This global, general-purpose geographic location encoder learns an implicit representation of locations by matching CNN and ViT inferred visual patterns of openly available satellite imagery with their geographic coordinates. The resulting SatCLIP location encoder efficiently summarizes the characteristics of any given location for convenient use in downstream tasks. In our experiments, we use SatCLIP embeddings to improve performance on nine diverse geospatial prediction tasks including temperature prediction, animal recognition, and population density estimation. Across tasks, SatCLIP consistently outperforms alternative location encoders and shows promise for improving geographic domain adaptation. These results demonstrate the potential of vision-location models to learn meaningful representations of our planet from the vast, varied, and largely untapped modalities of geospatial data.
DiffuseHigh: Training-Free Progressive High-Resolution Image Synthesis Through Structure Guidance
PDF ↗Large-scale generative models, such as text-to-image diffusion models, have garnered widespread attention across diverse domains due to their creative and high-fidelity image generation. Nonetheless, existing large-scale diffusion models are confined to generating images of up to 1K resolution, which is far from meeting the demands of contemporary commercial applications. Directly sampling higher-resolution images often yields results marred by artifacts such as object repetition and distorted shapes. Addressing the aforementioned issues typically necessitates training or fine-tuning models on higher-resolution datasets. However, this poses a formidable challenge due to the difficulty in collecting large-scale high-resolution images and substantial computational resources. While several preceding works have proposed alternatives to bypass the cumbersome training process, they often fail to produce convincing results. In this work, we probe the generative ability of diffusion models at higher resolution beyond their original capability and propose a novel progressive approach that fully utilizes generated low-resolution images to guide the generation of higher-resolution images. Additionally, we integrate an image sharpening operation into our pipeline, further enhancing image quality. Our method obviates the need for additional training or fine-tuning which significantly lowers the burden of computational costs. Extensive experiments and results validate the efficiency and efficacy of our method.
Multi-Modal Grounded Planning and Efficient Replanning for Learning Embodied Agents with a Few Examples
PDF ↗Learning a perception and reasoning module for robotic assistants to plan steps to perform complex tasks based on natural language instructions often requires large free-form language annotations, especially for short high-level instructions. To reduce the cost of annotation, large language models (LLMs) are used as a planner with few data. However, when elaborating the steps, even the state-of-the-art planner that uses LLMs mostly relies on linguistic common sense, often neglecting the status of the environment at command reception, resulting in inappropriate plans. To generate plans grounded in the environment, we propose FLARE (Few-shot Language with environmental Adaptive Replanning Embodied agent), which improves task planning using both language command and environmental perception. As language instructions often contain ambiguities or incorrect expressions, we additionally propose to correct the mistakes using visual cues from the agent. The proposed scheme allows us to use a few language pairs thanks to the visual cues and outperforms state-of-the-art approaches. Our code and the dataset are publicly available to facilitate further research.
Recent lightweight image captioning models using retrieved data mainly focus on text prompts. However, previous works only utilize the retrieved text as text prompts, and the visual information relies only on the CLIP visual embedding. Because of this issue, there is a limitation that the image descriptions inherent in the prompt are not sufficiently reflected in the visual embedding space. To tackle this issue, we propose ViPCap, a novel retrieval text-based visual prompt for lightweight image captioning. ViPCap leverages the retrieved text with image information as visual prompts to enhance the ability of the model to capture relevant visual information. By mapping text prompts into the CLIP space and generating multiple randomized Gaussian distributions, our method leverages sampling to explore randomly augmented distributions and effectively retrieves the semantic features that contain image information. These retrieved features are integrated into the image and designated as the visual prompt, leading to performance improvements on the datasets such as COCO, Flickr30k, and NoCaps. Experimental results demonstrate that ViPCap significantly outperforms prior lightweight captioning models in efficiency and effectiveness, demonstrating the potential for a plug-and-play solution.
This paper introduces a method for efficiently interpolating 3D dynamic sequences using truncated signed distance function (TSDF) volumes. The method calculates bi-directional motions between TSDF volumes of two frames and refines them to reconstruct intermediate frames. Unlike point cloud-based methods, which can suffer from varying and irregular point densities, the uniform and dense grid structure of TSDF offers a consistent framework for estimating the true motion of objects within a scene. In our experiments, the TSDF-based method offers more precise and reliable smooth motion prediction compared to the often error-prone surface depiction in point clouds. Experimental results demonstrate improved accuracy and reduced computational complexity, making it suitable for real-time applications.
Conventional GAN-based models for talking head generation often suffer from limited quality and unstable training. Recent approaches based on diffusion models have attempted to address these limitations and improve fidelity. However, they still face challenges, such as intensive sampling times and difficulties in maintaining temporal consistency due to the high stochasticity of diffusion models. To overcome these challenges, we propose a novel motion-disentangled diffusion model for high-quality talking head generation, called MoDiTalker. We introduce two modules: the Audio-To-Motion (AToM) module, designed to generate synchronized lip movements from audio, and the Motion-To-Video (MToV) module, designed to produce high-quality talking head videos based on the generated motions. AToM excels in capturing subtle lip movements by leveraging an audio attention mechanism. Additionally, MToV enhances temporal consistency by utilizing an efficient tri-plane representation. Our experiments on standard benchmarks demonstrate that our model outperforms existing GAN-based and diffusion-based models. We also provide comprehensive ablation studies and user study results.
With the growing demand for solutions to real-world video challenges, interest in dense video captioning (DVC) has been on the rise. DVC involves the automatic captioning and localization of untrimmed videos. Several studies highlight the challenges of DVC and introduce improved methods utilizing prior knowledge such as pre-training and external memory. In this research, we propose a model that leverages the prior knowledge of human-oriented hierarchical dense memory inspired by human memory hierarchy and cognition. To mimic human-like memory recall, we construct a hierarchical memory and a hierarchical memory reading module. We build an efficient hierarchical dense memory by employing clustering of memory events and summarization using large language models. Comparative experiments demonstrate that this hierarchical memory recall process improves the performance of DVC by achieving state-of-the-art performance on YouCook2 and ViTT datasets.
ProtoOcc: Accurate, Efficient 3D Occupancy Prediction Using Dual Branch Encoder-Prototype Query Decoder
PDF ↗In this paper, we introduce ProtoOcc, a novel 3D occupancy prediction model designed to predict the occupancy states and semantic classes of 3D voxels via a deep semantic understanding of scenes. ProtoOcc consists of two main components: the Dual Branch Encoder (DBE) and the Prototype Query Decoder (PQD). The DBE produces a new 3D voxel representation by combining 3D voxel and BEV representations across multiple scales using a dual branch structure. This design combines the BEV representation, which offers a large receptive field, with the voxel representation, known for its higher spatial resolution, thereby improving both performance and computational efficiency. The PQD employs two types of prototype-based queries to expedite the Transformer decoding process. Scene-Adaptive Prototypes are generated from the 3D voxel features of the input sample, while Scene-Agnostic Prototypes are updated during training using an Exponential Moving Average of the Scene-Adaptive Prototypes. Using these prototype-based queries for decoding, we can directly predict 3D occupancy in a single step, eliminating the need for iterative Transformer decoding. Additionally, we propose Robust Prototype Learning, which introduces noise into the prototype generation process and trains the model to denoise during the training phase. This approach enhances the robustness of ProtoOcc against degraded prototype feature quality. ProtoOcc achieves state-of-the-art performance with 45.02% mIoU on the Occ3D-nuScenes benchmark. For the single-frame method, it reaches 39.56% mIoU with 12.83 FPS on an NVIDIA RTX 3090.
Speech-driven 3D facial animation has garnered lots of attention thanks to its broad range of applications. Despite recent advancements in achieving realistic lip motion, current methods fail to capture the nuanced emotional undertones conveyed through speech and produce monotonous facial motion. These limitations result in blunt and repetitive facial animations, reducing user engagement and hindering their applicability. To address these challenges, we introduce DEEPTalk, a novel approach that generates diverse and emotionally rich 3D facial expressions directly from speech inputs. To achieve this, we first train DEE (Dynamic Emotion Embedding), which employs probabilistic contrastive learning to forge a joint emotion embedding space for both speech and facial motion. This probabilistic framework captures the uncertainty in interpreting emotions from speech and facial motion, enabling the derivation of emotion vectors from its multifaceted space. Moreover, to generate dynamic facial motion, we design TH-VQVAE (Temporally Hierarchical VQ-VAE) as an expressive and robust motion prior overcoming limitations of VAEs and VQ-VAEs. Utilizing these strong priors, we develop DEEPTalk, a talking head generator that non-autoregressively predicts codebook indices to create dynamic facial motion, incorporating a novel emotion consistency loss. Extensive experiments on various datasets demonstrate the effectiveness of our approach in creating diverse, emotionally expressive talking faces that maintain accurate lip-sync. Our project page is available at https://whwjdqls.github.io/deeptalk.github.io/.
Temporal Action Detection (TAD) is fundamental yet challenging for real-world video applications. Leveraging the unique benefits of transformers, various DETR-based approaches have been adopted in TAD. However, it has recently been identified that the attention collapse in self-attention causes the performance degradation of DETR for TAD. Building upon previous research, this paper newly addresses the attention collapse problem in cross-attention within DETR-based TAD methods. Moreover, our findings reveal that cross-attention exhibits patterns distinct from predictions, indicating a short-cut phenomenon. To resolve this, we propose a new framework, Prediction-Feedback DETR (Pred-DETR), which utilizes predictions to restore the collapse and align the cross- and self-attention with predictions. Specifically, we devise novel prediction-feedback objectives using guidance from the relations of the predictions. As a result, Pred-DETR significantly alleviates the collapse and achieves state-of-the-art performance among DETR-based methods on various challenging benchmarks including THUMOS14, ActivityNet-v1.3, HACS, and FineAction.
Recent advancements in self-supervised denoising have made it possible to train models without needing a large amount of noisy-clean image pairs. A significant development in this area is the use of blind-spot networks (BSNs), which use single noisy images as training pairs by masking some input information to prevent noise transmission to the network output. Researchers have shown that BSNs are capable of reconstructing clean pixels from various types of independent pixel-wise degradations, such as synthetic additive white Gaussian noise (AWGN). However, unlike synthetic noise, real noise often contains highly correlated components which can induce noise transmission and reduce the performance of BSNs. To address the spatial correlation of real noise, we propose the Adjacent Pixel Replacer (APR), which decorrelates noise without a downsampling process that is widely adopted in previous research. The dissimilarity in our APR-generated pairs serves as relatively different noise components during training. Hence, it enables the BSN to block noise transmission while utilizing clean information effectively. As a result, BSN can utilize denser information to reconstruct the corresponding center pixel. We also propose Recharged Distillation (RD) to enhance high-frequency textures without additional network modifications. This method selectively refines clean information from recharged noisy pixels during distillation. Extensive experimental results demonstrate that our proposed method outperforms the existing state-of-the-art self-supervised denoising methods in real sRGB space.
Generalized Zero-Shot Learning for Point Cloud Segmentation with Evidence-Based Dynamic Calibration
PDF ↗Generalized zero-shot semantic segmentation of 3D point clouds aims to classify each point into both seen and unseen classes. A significant challenge with these models is their tendency to make biased predictions, often favoring the classes encountered during training. This problem is more pronounced in 3D applications, where the scale of the training data is typically smaller than in image-based tasks. To address this problem, we propose a novel method called E3DPC-GZSL, which reduces overconfident predictions towards seen classes without relying on separate classifiers for seen and unseen data. E3DPC-GZSL tackles the overconfidence problem by integrating an evidence-based uncertainty estimator into a classifier. This estimator is then used to adjust prediction probabilities using a dynamic calibrated stacking factor that accounts for pointwise prediction uncertainty. In addition, E3DPC-GZSL introduces a novel training strategy that improves uncertainty estimation by refining the semantic space. This is achieved by merging learnable parameters with text-derived features, thereby improving model optimization for unseen data. Extensive experiments demonstrate that the proposed approach achieves state-of-the-art performance on generalized zero-shot semantic segmentation datasets, including ScanNet v2 and S3DIS.
3D point clouds are increasingly vital for applications like autonomous driving and robotics, yet the raw data captured by sensors often suffer from noise and sparsity, creating challenges for downstream tasks. Consequently, point cloud upsampling becomes essential for improving density and uniformity, with recent approaches showing promise by projecting randomly generated query points onto the underlying surface of sparse point clouds. However, these methods often result in outliers, non-uniformity, and difficulties in handling regions with high curvature and intricate structures. In this work, we address these challenges by introducing the Progressive Local Surface Estimator (PLSE), which more effectively captures local features in complex regions through a curvature-based sampling technique that selectively targets high-curvature areas. Additionally, we incorporate a curriculum learning strategy that leverages the curvature distribution within the point cloud to naturally assess the sample difficulty, enabling curriculum learning on point cloud data for the first time. The experimental results demonstrate that our approach significantly outperforms existing methods, achieving high-quality, dense point clouds with superior accuracy and detail.
Foundational vision-language models like CLIP are emerging as a promising paradigm in vision due to their excellent generalization. However, adapting these models for downstream tasks while maintaining their generalization remains challenging. In literature, one branch of methods adapts CLIP by learning prompts using images. While effective, these methods often rely on image-label data, which is not always practical, and struggle to generalize to new datasets due to overfitting on few-shot source data. Another approach explores training-free methods by generating class captions from large language models (LLMs) and performing prompt ensembling, but these methods often produce static, class-specific prompts that cannot be transferred to new classes and incur additional costs by generating LLM descriptions for each class separately. In this work, we aim to combine the strengths of both approaches by learning prompts using only text data derived from LLMs. As supervised training of prompts in the image-free setup is non-trivial, we develop a language-only efficient training approach that enables prompts to distill rich contextual knowledge from LLM data. Furthermore, by mapping the LLM contextual text data within the learned prompts, our approach enables zero-shot transfer of prompts to new classes and datasets, potentially reducing the LLM prompt engineering cost. To the best of our knowledge, this is the first work that learns generalized and transferable prompts for image tasks using only text data. We perform evaluations on 4 benchmarks, where ProText improves over ensembling methods while being competitive with those using labeled images.
The objective of referring expression comprehension (REC) is to accurately identify the object in an image described by a given expression. Existing REC methods, including transformer-based and graph-based approaches among others, have shown robust performance in REC tasks. In this study, we present a groundbreaking framework named DiffusionREC for REC task. This framework reimagines the REC task as a text guided bounding box denoising diffusion process, through which noisy bounding boxes are refined and distilled to pinpoint the target box. Throughout the training process, the bounding box of the target object diffuses from its ground-truth position towards a random distribution. Simultaneously, a filtering-based object decoder is introduced to reverse this diffusion of noise, conditional on the provided expression, the result from previous denoised step and the interaction between the expression and the image. At the inference stage, we begin by randomly generating a collection of boxes. Subsequently, the filtering-based object decoder is iteratively employed to refine and prune these bounding boxes, taking into account the conditions on the given expression, the results from the previous denoised step, and the interaction between the expression and the image. Extensive experiments conducted on six datasets demonstrate that DiffusionREC outperforms previous REC methods, yielding superior performances.
Guided depth super-resolution (GDSR) has demonstrated impressive performance across a wide range of domains, with numerous methods being proposed. However, existing methods often treat depth maps as images, where shading values are computed discretely, making them struggle to effectively restore the continuity inherent in the depth map. In this paper, we propose a novel approach that maximizes the utilization of spatial characteristics in depth, coupled with human abstract perception of real-world substance, by transforming the GDSR issue into deformation of a roughcast with ideal plasticity, which can be deformed by force like a continuous object. Specifically, we firstly designed a cross-modal operation, Continuity-constrained Asymmetrical Pixelwise Operation (CAPO), which can mimic the process of deforming an isovolumetrically flexible object through external forces. Utilizing CAPO as the fundamental component, we develop the Pixelwise Cross Gradient Deformation (PCGD), which is capable of emulating operations on ideal plastic objects (without volume constraint). Notably, our approach demonstrates state-of-the-art performance across four widely adopted benchmarks for GDSR, with significant advantages in large-scale tasks and generalizability.
Neural Radiance Fields (NeRF) have achieved huge success in effectively capturing and representing 3D objects and scenes. However, to establish an ubiquitous presence in everyday media formats, such as images and videos, we need to fulfill three key objectives: 1. fast encoding and decoding time, 2. compact model sizes, and 3. high-quality renderings. Despite recent advancements, a comprehensive algorithm that adequately addresses all objectives has yet to be fully realized. In this work, we present CodecNeRF, a neural codec for NeRF representations, consisting of an encoder and decoder architecture that can generate a NeRF representation in a single forward pass. Furthermore, inspired by the recent parameter-efficient finetuning approaches, we propose a finetuning method to efficiently adapt the generated NeRF representations to a new test instance, leading to high-quality image renderings and compact code sizes. The proposed CodecNeRF, a newly suggested encoding-decoding-finetuning pipeline for NeRF, achieved unprecedented compression performance of more than 100x and a remarkable reduction in encoding time while maintaining (or improving) the image quality on widely used 3D object datasets.
Contextual information at the video level has become increasingly crucial for visual object tracking. However, existing methods typically use only a few tokens to convey this information, which can lead to information loss and limit their ability to fully capture the context. To address this issue, we propose a new video-level visual object tracking framework called MCITrack. It leverages Mamba's hidden states to continuously record and transmit extensive contextual information throughout the video stream, resulting in more robust object tracking. The core component of MCITrack is the Contextual Information Fusion module, which consists of the mamba layer and the cross-attention layer. The mamba layer stores historical contextual information, while the cross-attention layer integrates this information into the current visual features of each backbone block. This module enhances the model's ability to capture and utilize contextual information at multiple levels through deep integration with the backbone. Experiments demonstrate that MCITrack achieves competitive performance across numerous benchmarks. For instance, it gets 76.6% AUC on LaSOT and 80.0% AO on GOT-10k, establishing a new state-of-the-art performance.