Xiaofei He 0001

ICLR Conference 2025 Conference Paper

Depth Any Video with Scalable Synthetic Data

• Honghui Yang
• Di Huang
• Wei Yin 0006
• Chunhua Shen
• Haifeng Liu 0001
• Xiaofei He 0001
• Binbin Lin
• Wanli Ouyang

Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse virtual environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates—even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency. The code and model weights are open-sourced.

ICLR Conference 2025 Conference Paper

Improving Complex Reasoning with Dynamic Prompt Corruption: A Soft Prompt Optimization Approach

• Sinan Fan
• Liang Xie 0003
• Chen Shen 0003
• Ge Teng
• Xiaosong Yuan
• Xiaofeng Zhang 0006
• Chenxi Huang 0004
• Wenxiao Wang 0001

Prompt Tuning (PT) has emerged as a promising Parameter-Efficient Fine-Tuning (PEFT) approach by appending trainable continuous prompt vectors to the input, maintaining competitive performance with significantly fewer trainable parameters. While PT has shown effectiveness in enhancing task performance, particularly for classification tasks, its application to complex reasoning tasks has been largely overlooked. Our investigation reveals that PT provides limited improvement and may even degrade performance in reasoning tasks. This phenomenon suggests that soft prompts can positively impact certain instances while negatively affecting others, particularly during the latter stages of reasoning. To address these challenges, we propose a novel method called Dynamic Prompt Corruption (DPC), which seeks to optimize the use of soft prompts in reasoning tasks. DPC dynamically adjusts the influence of soft prompts based on their impact on the reasoning process. Specifically, it involves two key components: Dynamic Trigger and Dynamic Corruption. Dynamic Trigger measures the influence of soft prompts, determining whether their impact is beneficial or detrimental. Dynamic Corruption mitigates the negative effects of soft prompts by selectively masking key tokens that interfere with the reasoning process. We validate our approach through extensive experiments on various large language models (LLMs) and reasoning tasks, including GSM8K, MATH, and AQuA. The results demonstrate that Dynamic Prompt Corruption consistently improves the performance of LLMs, achieving 4\%-8\% accuracy gains compared to standard prompt tuning. These findings highlight the effectiveness of our approach and its potential to enhance complex reasoning in LLMs.

ICLR Conference 2025 Conference Paper

SAMRefiner: Taming Segment Anything Model for Universal Mask Refinement

• Yuqi Lin
• Hengjia Li
• Wenqi Shao
• Zheng Yang 0008
• Jun Zhao 0009
• Xiaofei He 0001
• Ping Luo 0002
• Kaipeng Zhang

In this paper, we explore a principal way to enhance the quality of widely pre-existing coarse masks, enabling them to serve as reliable training data for segmentation models to reduce the annotation cost. In contrast to prior refinement techniques that are tailored to specific models or tasks in a close-world manner, we propose SAMRefiner, a universal and efficient approach by adapting SAM to the mask refinement task. The core technique of our model is the noise-tolerant prompting scheme. Specifically, we introduce a multi-prompt excavation strategy to mine diverse input prompts for SAM (\ie, distance-guided points, context-aware elastic bounding boxes, and Gaussian-style masks) from initial coarse masks. These prompts can collaborate with each other to mitigate the effect of defects in coarse masks. In particular, considering the difficulty of SAM to handle the multi-object case in semantic segmentation, we introduce a split-then-merge (STM) pipeline. Additionally, we extend our method to SAMRefiner++ by introducing an additional IoU adaption step to further boost the performance of the generic SAMRefiner on the target dataset. This step is self-boosted and requires no additional annotation. The proposed framework is versatile and can flexibly cooperate with existing segmentation methods. We evaluate our mask framework on a wide range of benchmarks under different settings, demonstrating better accuracy and efficiency. SAMRefiner holds significant potential to expedite the evolution of refinement tools. Our code is available at https://github.com/linyq2117/SAMRefiner.

ICLR Conference 2024 Conference Paper

Few-shot Hybrid Domain Adaptation of Image Generator

• Hengjia Li
• Yang Liu 0212
• Linxuan Xia
• Yuqi Lin
• Wenxiao Wang 0001
• Tu Zheng
• Zheng Yang 0008
• Xiaohui Zhong

Can a pre-trained generator be adapted to the hybrid of multiple target domains and generate images with integrated attributes of them? In this work, we introduce a new task -- Few-shot $\textit{Hybrid Domain Adaptation}$ (HDA). Given a source generator and several target domains, HDA aims to acquire an adapted generator that preserves the integrated attributes of all target domains, without overriding the source domain's characteristics. Compared with $\textit{Domain Adaptation}$ (DA), HDA offers greater flexibility and versatility to adapt generators to more composite and expansive domains. Simultaneously, HDA also presents more challenges than DA as we have access only to images from individual target domains and lack authentic images from the hybrid domain. To address this issue, we introduce a discriminator-free framework that directly encodes different domains' images into well-separable subspaces. To achieve HDA, we propose a novel directional subspace loss comprised of a distance loss and a direction loss. Concretely, the distance loss blends the attributes of all target domains by reducing the distances from generated images to all target subspaces. The direction loss preserves the characteristics from the source domain by guiding the adaptation along the perpendicular to subspaces. Experiments show that our method can obtain numerous domain-specific attributes in a single adapted generator, which surpasses the baseline methods in semantic similarity, image fidelity, and cross-domain consistency.

ICRA Conference 2023 Conference Paper

PriorLane: A Prior Knowledge Enhanced Lane Detection Approach Based on Transformer

• Qibo Qiu
• Haiming Gao
• Wei Hua 0002
• Gang Huang 0004
• Xiaofei He 0001

Lane detection is one of the fundamental modules in self-driving. In this paper we employ a transformer-only method for lane detection, thus it could benefit from the blooming development of fully vision transformer and achieve the state-of-the-art (SOTA) performance on both CULane and TuSimple benchmarks, by fine-tuning the weight fully pre-trained on large datasets. More importantly, this paper proposes a novel and general framework called PriorLane, which is used to enhance the segmentation performance of the fully vision transformer by introducing the low-cost local prior knowledge. Specifically, PriorLane utilizes an encoder-only transformer to fuse the feature extracted by a pre-trained segmentation model with prior knowledge embeddings. Note that a Knowledge Embedding Alignment (KEA) module is adapted to enhance the fusion performance by aligning the knowledge embedding. Extensive experiments on our Zjlab dataset show that PriorLane outperforms SOTA lane detection methods by a 2. 82% mIoU when prior knowledge is employed, and the code will be released at: https://github.com/vincentqqb/PriorLane.

ICLR Conference 2022 Conference Paper

CrossFormer: A Versatile Vision Transformer Hinging on Cross-scale Attention

• Wenxiao Wang 0001
• Lu Yao
• Long Chen 0016
• Binbin Lin
• Deng Cai 0001
• Xiaofei He 0001
• Wei Liu 0005

Transformers have made great progress in dealing with computer vision tasks. However, existing vision transformers have not yet possessed the ability of building the interactions among features of different scales, which is perceptually important to visual inputs. The reasons are two-fold: (1) Input embeddings of each layer are equal-scale, so no cross-scale feature can be extracted; (2) to lower the computational cost, some vision transformers merge adjacent embeddings inside the self-attention module, thus sacrificing small-scale (fine-grained) features of the embeddings and also disabling the cross-scale interactions. To this end, we propose Cross-scale Embedding Layer (CEL) and Long Short Distance Attention (LSDA). On the one hand, CEL blends each embedding with multiple patches of different scales, providing the self-attention module itself with cross-scale features. On the other hand, LSDA splits the self-attention module into a short-distance one and a long-distance counterpart, which not only reduces the computational burden but also keeps both small-scale and large-scale features in the embeddings. Through the above two designs, we achieve cross-scale attention. Besides, we put forward a dynamic position bias for vision transformers to make the popular relative position bias apply to variable-sized images. Hinging on the cross-scale attention module, we construct a versatile vision architecture, dubbed CrossFormer, which accommodates variable-sized inputs. Extensive experiments show that CrossFormer outperforms the other vision transformers on image classification, object detection, instance segmentation, and semantic segmentation tasks.

ICLR Conference 2022 Conference Paper

WeakM3D: Towards Weakly Supervised Monocular 3D Object Detection

• Liang Peng
• Senbo Yan
• Boxi Wu
• Zheng Yang 0008
• Xiaofei He 0001
• Deng Cai 0001

Monocular 3D object detection is one of the most challenging tasks in 3D scene understanding. Due to the ill-posed nature of monocular imagery, existing monocular 3D detection methods highly rely on training with the manually annotated 3D box labels on the LiDAR point clouds. This annotation process is very laborious and expensive. To dispense with the reliance on 3D box labels, in this paper we explore the weakly supervised monocular 3D detection. Specifically, we first detect 2D boxes on the image. Then, we adopt the generated 2D boxes to select corresponding RoI LiDAR points as the weak supervision. Eventually, we adopt a network to predict 3D boxes which can tightly align with associated RoI LiDAR points. This network is learned by minimizing our newly-proposed 3D alignment loss between the 3D box estimates and the corresponding RoI LiDAR points. We will illustrate the potential challenges of the above learning problem and resolve these challenges by introducing several effective designs into our method. Codes are available at https://github.com/SPengLiang/WeakM3D.

ICML Conference 2021 Conference Paper

Accelerate CNNs from Three Dimensions: A Comprehensive Pruning Framework

• Wenxiao Wang 0001
• Minghao Chen 0001
• Shuai Zhao 0006
• Long Chen 0016
• Jinming Hu
• Haifeng Liu 0001
• Deng Cai 0001
• Xiaofei He 0001

Most neural network pruning methods, such as filter-level and layer-level prunings, prune the network model along one dimension (depth, width, or resolution) solely to meet a computational budget. However, such a pruning policy often leads to excessive reduction of that dimension, thus inducing a huge accuracy loss. To alleviate this issue, we argue that pruning should be conducted along three dimensions comprehensively. For this purpose, our pruning framework formulates pruning as an optimization problem. Specifically, it first casts the relationships between a certain model’s accuracy and depth/width/resolution into a polynomial regression and then maximizes the polynomial to acquire the optimal values for the three dimensions. Finally, the model is pruned along the three optimal dimensions accordingly. In this framework, since collecting too much data for training the regression is very time-costly, we propose two approaches to lower the cost: 1) specializing the polynomial to ensure an accurate regression even with less training data; 2) employing iterative pruning and fine-tuning to collect the data faster. Extensive experiments show that our proposed algorithm surpasses state-of-the-art pruning algorithms and even neural architecture search-based algorithms.

ICML Conference 2017 Conference Paper

Scaling Up Sparse Support Vector Machines by Simultaneous Feature and Sample Reduction

• Weizhong Zhang
• Bin Hong
• Wei Liu 0005
• Jieping Ye
• Deng Cai 0001
• Xiaofei He 0001
• Jie Wang 0005

Sparse support vector machine (SVM) is a popular classification technique that can simultaneously learn a small set of the most interpretable features and identify the support vectors. It has achieved great successes in many real-world applications. However, for large-scale problems involving a huge number of samples and extremely high-dimensional features, solving sparse SVMs remains challenging. By noting that sparse SVMs induce sparsities in both feature and sample spaces, we propose a novel approach, which is based on accurate estimations of the primal and dual optima of sparse SVMs, to simultaneously identify the features and samples that are guaranteed to be irrelevant to the outputs. Thus, we can remove the identified inactive samples and features from the training phase, leading to substantial savings in both the memory usage and computational cost without sacrificing accuracy. To the best of our knowledge, the proposed method is the first static feature and sample reduction method for sparse SVMs. Experiments on both synthetic and real datasets (e. g. , the kddb dataset with about 20 million samples and 30 million features) demonstrate that our approach significantly outperforms state-of-the-art methods and the speedup gained by our approach can be orders of magnitude.

ICML Conference 2014 Conference Paper

Geodesic Distance Function Learning via Heat Flow on Vector Fields

• Binbin Lin
• Ji Yang
• Xiaofei He 0001
• Jieping Ye

Learning a distance function or metric on a given data manifold is of great importance in machine learning and pattern recognition. Many of the previous works first embed the manifold to Euclidean space and then learn the distance function. However, such a scheme might not faithfully preserve the distance function if the original manifold is not Euclidean. In this paper, we propose to learn the distance function directly on the manifold without embedding. We first provide a theoretical characterization of the distance function by its gradient field. Based on our theoretical analysis, we propose to first learn the gradient field of the distance function and then learn the distance function itself. Specifically, we set the gradient field of a local distance function as an initial vector field. Then we transport it to the whole manifold via heat flow on vector fields. Finally, the geodesic distance function can be obtained by requiring its gradient field to be close to the normalized vector field. Experimental results on both synthetic and real data demonstrate the effectiveness of our proposed algorithm.

ICML Conference 2013 Conference Paper

O(logT) Projections for Stochastic Optimization of Smooth and Strongly Convex Functions

• Lijun Zhang 0005
• Tianbao Yang
• Rong Jin 0001
• Xiaofei He 0001

Traditional algorithms for stochastic optimization require projecting the solution at each iteration into a given domain to ensure its feasibility. When facing complex domains, such as the positive semidefinite cone, the projection operation can be expensive, leading to a high computational cost per iteration. In this paper, we present a novel algorithm that aims to reduce the number of projections for stochastic optimization. The proposed algorithm combines the strength of several recent developments in stochastic optimization, including mini-batches, extra-gradient, and epoch gradient descent, in order to effectively explore the smoothness and strong convexity. We show, both in expectation and with a high probability, that when the objective function is both smooth and strongly convex, the proposed algorithm achieves the optimal O(1/T) rate of convergence with only O(logT) projections. Our empirical study verifies the theoretical result.

ICML Conference 2013 Conference Paper

Online Kernel Learning with a Near Optimal Sparsity Bound

• Lijun Zhang 0005
• Jinfeng Yi
• Rong Jin 0001
• Ming Lin 0002
• Xiaofei He 0001

In this work, we focus on Online Sparse Kernel Learning that aims to online learn a kernel classifier with a bounded number of support vectors. Although many online learning algorithms have been proposed to learn a sparse kernel classifier, most of them fail to bound the number of support vectors used by the final solution which is the average of the intermediate kernel classifiers generated by online algorithms. The key idea of the proposed algorithm is to measure the difficulty in correctly classifying a training example by the derivative of a smooth loss function, and give a more chance to a difficult example to be a support vector than an easy one via a sampling scheme. Our analysis shows that when the loss function is smooth, the proposed algorithm yields similar performance guarantee as the standard online learning algorithm but with a near optimal number of support vectors (up to a poly(lnT) factor). Our empirical study shows promising performance of the proposed algorithm compared to the state-of-the-art algorithms for online sparse kernel learning.

ICML Conference 2009 Conference Paper

Probabilistic dyadic data analysis with local and global consistency

• Deng Cai 0001
• Xuanhui Wang
• Xiaofei He 0001

Dyadic data arises in many real world applications such as social network analysis and information retrieval. In order to discover the underlying or hidden structure in the dyadic data, many topic modeling techniques were proposed. The typical algorithms include Probabilistic Latent Semantic Analysis (PLSA) and Latent Dirichlet Allocation (LDA). The probability density functions obtained by both of these two algorithms are supported on the Euclidean space. However, many previous studies have shown naturally occurring data may reside on or close to an underlying submanifold. We introduce a probabilistic framework for modeling both the topical and geometrical structure of the dyadic data that explicitly takes into account the local manifold structure. Specifically, the local manifold structure is modeled by a graph. The graph Laplacian, analogous to the Laplace-Beltrami operator on manifolds, is applied to smooth the probability density functions. As a result, the obtained probabilistic distributions are concentrated around the data manifold. Experimental results on real data sets demonstrate the effectiveness of the proposed approach.

ICML Conference 2005 Conference Paper

Statistical and computational analysis of locality preserving projection

• Xiaofei He 0001
• Deng Cai 0001
• Wanli Min