Sinisa Todorovic

Khoi Nguyen, Sinisa Todorovic

This paper addresses incremental few-shot instance segmentation, where a few examples of new object classes arrive when access to training examples of old classes is not available anymore, and the goal is to perform well on both old and new classes. We make two contributions by extending the common Mask-RCNN framework in its second stage -- namely, we specify a new object class classifier based on the probit function and a new uncertainty-guided bounding-box predictor. The former leverages Bayesian learning to address a paucity of training examples of new classes. The latter learns not only to predict object bounding boxes but also to estimate the uncertainty of the prediction as guidance for bounding box refinement. We also specify two new loss functions in terms of the estimated object-class distribution and bounding-box uncertainty. Our contributions produce significant performance gains on the COCO dataset over the state of the art -- specifically, the gain of +6 on the new classes and +16 on the old classes in the AP instance segmentation metric. Furthermore, we are the first to evaluate the incremental few-shot setting on the more challenging LVIS dataset.

Nikola Andrejic, Milica Spasic, Igor Mihajlovic et al.

This work introduces Ui2i, a novel model for unpaired image-to-image translation, trained on content-wise unpaired datasets to enable style transfer across domains while preserving content. Building on CycleGAN, Ui2i incorporates key modifications to better disentangle content and style features, and preserve content integrity. Specifically, Ui2i employs U-Net-based generators with skip connections to propagate localized shallow features deep into the generator. Ui2i removes feature-based normalization layers from all modules and replaces them with approximate bidirectional spectral normalization -- a parameter-based alternative that enhances training stability. To further support content preservation, channel and spatial attention mechanisms are integrated into the generators. Training is facilitated through image scale augmentation. Evaluation on two biomedical tasks -- domain adaptation for nuclear segmentation in immunohistochemistry (IHC) images and unmixing of biological structures superimposed in single-channel immunofluorescence (IF) images -- demonstrates Ui2i's ability to preserve content fidelity in settings that demand more accurate structural preservation than typical translation tasks. To the best of our knowledge, Ui2i is the first approach capable of separating superimposed signals in IF images using real, unpaired training data.

Tieqiao Wang, Sinisa Todorovic

Most recent work on action segmentation relies on pre-computed frame features from models trained on other tasks and typically focuses on framewise encoding and labeling without explicitly modeling action segments. To overcome these limitations, we introduce the End-to-End Action Segmentation Transformer (EAST), which processes raw video frames directly -- eliminating the need for pre-extracted features and enabling true end-to-end training. Our contributions are as follows: (1) a lightweight adapter design for effective fine-tuning of large backbones; (2) an efficient segmentation-by-detection framework for leveraging action proposals predicted over a coarsely downsampled video; and (3) a novel action-proposal-based data augmentation strategy. EAST achieves SOTA performance on standard benchmarks, including GTEA, 50Salads, Breakfast, and Assembly-101.

Arjun R. Akula, Keze Wang, Changsong Liu et al.

We propose CX-ToM, short for counterfactual explanations with theory-of mind, a new explainable AI (XAI) framework for explaining decisions made by a deep convolutional neural network (CNN). In contrast to the current methods in XAI that generate explanations as a single shot response, we pose explanation as an iterative communication process, i.e. dialog, between the machine and human user. More concretely, our CX-ToM framework generates sequence of explanations in a dialog by mediating the differences between the minds of machine and human user. To do this, we use Theory of Mind (ToM) which helps us in explicitly modeling human's intention, machine's mind as inferred by the human as well as human's mind as inferred by the machine. Moreover, most state-of-the-art XAI frameworks provide attention (or heat map) based explanations. In our work, we show that these attention based explanations are not sufficient for increasing human trust in the underlying CNN model. In CX-ToM, we instead use counterfactual explanations called fault-lines which we define as follows: given an input image I for which a CNN classification model M predicts class c_pred, a fault-line identifies the minimal semantic-level features (e.g., stripes on zebra, pointed ears of dog), referred to as explainable concepts, that need to be added to or deleted from I in order to alter the classification category of I by M to another specified class c_alt. We argue that, due to the iterative, conceptual and counterfactual nature of CX-ToM explanations, our framework is practical and more natural for both expert and non-expert users to understand the internal workings of complex deep learning models. Extensive quantitative and qualitative experiments verify our hypotheses, demonstrating that our CX-ToM significantly outperforms the state-of-the-art explainable AI models.

Khoi Nguyen, Sinisa Todorovic

This paper addresses weakly supervised amodal instance segmentation, where the goal is to segment both visible and occluded (amodal) object parts, while training provides only ground-truth visible (modal) segmentations. Following prior work, we use data manipulation to generate occlusions in training images and thus train a segmenter to predict amodal segmentations of the manipulated data. The resulting predictions on training images are taken as the pseudo-ground truth for the standard training of Mask-RCNN, which we use for amodal instance segmentation of test images. For generating the pseudo-ground truth, we specify a new Amodal Segmenter based on Boundary Uncertainty estimation (ASBU) and make two contributions. First, while prior work uses the occluder's mask, our ASBU uses the occlusion boundary as input. Second, ASBU estimates an uncertainty map of the prediction. The estimated uncertainty regularizes learning such that lower segmentation loss is incurred on regions with high uncertainty. ASBU achieves significant performance improvement relative to the state of the art on the COCOA and KINS datasets in three tasks: amodal instance segmentation, amodal completion, and ordering recovery.

Jun Li, Sinisa Todorovic

This paper addresses unsupervised action segmentation. Prior work captures the frame-level temporal structure of videos by a feature embedding that encodes time locations of frames in the video. We advance prior work with a new self-supervised learning (SSL) of a feature embedding that accounts for both frame- and action-level structure of videos. Our SSL trains an RNN to recognize positive and negative action sequences, and the RNN's hidden layer is taken as our new action-level feature embedding. The positive and negative sequences consist of action segments sampled from videos, where in the former the sampled action segments respect their time ordering in the video, and in the latter they are shuffled. As supervision of actions is not available and our SSL requires access to action segments, we specify an HMM that explicitly models action lengths, and infer a MAP action segmentation with the Viterbi algorithm. The resulting action segmentation is used as pseudo-ground truth for estimating our action-level feature embedding and updating the HMM. We alternate the above steps within the Generalized EM framework, which ensures convergence. Our evaluation on the Breakfast, YouTube Instructions, and 50Salads datasets gives superior results to those of the state of the art.

Jun Li, Sinisa Todorovic

This paper is about action segmentation under weak supervision in training, where the ground truth provides only a set of actions present, but neither their temporal ordering nor when they occur in a training video. We use a Hidden Markov Model (HMM) grounded on a multilayer perceptron (MLP) to label video frames, and thus generate a pseudo-ground truth for the subsequent pseudo-supervised training. In testing, a Monte Carlo sampling of action sets seen in training is used to generate candidate temporal sequences of actions, and select the maximum posterior sequence. Our key contribution is a new anchor-constrained Viterbi algorithm (ACV) for generating the pseudo-ground truth, where anchors are salient action parts estimated for each action from a given ground-truth set. Our evaluation on the tasks of action segmentation and alignment on the benchmark Breakfast, MPII Cooking2, Hollywood Extended datasets demonstrates our superior performance relative to that of prior work.

Khoi Nguyen, Sinisa Todorovic

This paper is about few-shot instance segmentation, where training and test image sets do not share the same object classes. We specify and evaluate a new few-shot anchor-free part-based instance segmenter FAPIS. Our key novelty is in explicit modeling of latent object parts shared across training object classes, which is expected to facilitate our few-shot learning on new classes in testing. We specify a new anchor-free object detector aimed at scoring and regressing locations of foreground bounding boxes, as well as estimating relative importance of latent parts within each box. Also, we specify a new network for delineating and weighting latent parts for the final instance segmentation within every detected bounding box. Our evaluation on the benchmark COCO-20i dataset demonstrates that we significantly outperform the state of the art.

Khoi Nguyen, Sinisa Todorovic

This paper addresses unsupervised few-shot object recognition, where all training images are unlabeled, and test images are divided into queries and a few labeled support images per object class of interest. The training and test images do not share object classes. We extend the vanilla GAN with two loss functions, both aimed at self-supervised learning. The first is a reconstruction loss that enforces the discriminator to reconstruct the probabilistically sampled latent code which has been used for generating the "fake" image. The second is a triplet loss that enforces the discriminator to output image encodings that are closer for more similar images. Evaluation, comparisons, and detailed ablation studies are done in the context of few-shot classification. Our approach significantly outperforms the state of the art on the Mini-Imagenet and Tiered-Imagenet datasets.

Jun Li, Sinisa Todorovic

This paper is about weakly supervised action segmentation, where the ground truth specifies only a set of actions present in a training video, but not their true temporal ordering. Prior work typically uses a classifier that independently labels video frames for generating the pseudo ground truth, and multiple instance learning for training the classifier. We extend this framework by specifying an HMM, which accounts for co-occurrences of action classes and their temporal lengths, and by explicitly training the HMM on a Viterbi-based loss. Our first contribution is the formulation of a new set-constrained Viterbi algorithm (SCV). Given a video, the SCV generates the MAP action segmentation that satisfies the ground truth. This prediction is used as a framewise pseudo ground truth in our HMM training. Our second contribution in training is a new regularization of feature affinities between training videos that share the same action classes. Evaluation on action segmentation and alignment on the Breakfast, MPII Cooking2, Hollywood Extended datasets demonstrates our significant performance improvement for the two tasks over prior work.

Jun Li, Li Fuxin, Sinisa Todorovic

Strictly enforcing orthonormality constraints on parameter matrices has been shown advantageous in deep learning. This amounts to Riemannian optimization on the Stiefel manifold, which, however, is computationally expensive. To address this challenge, we present two main contributions: (1) A new efficient retraction map based on an iterative Cayley transform for optimization updates, and (2) An implicit vector transport mechanism based on the combination of a projection of the momentum and the Cayley transform on the Stiefel manifold. We specify two new optimization algorithms: Cayley SGD with momentum, and Cayley ADAM on the Stiefel manifold. Convergence of Cayley SGD is theoretically analyzed. Our experiments for CNN training demonstrate that both algorithms: (a) Use less running time per iteration relative to existing approaches that enforce orthonormality of CNN parameters; and (b) Achieve faster convergence rates than the baseline SGD and ADAM algorithms without compromising the performance of the CNN. Cayley SGD and Cayley ADAM are also shown to reduce the training time for optimizing the unitary transition matrices in RNNs.

Jun Li, Peng Lei, Sinisa Todorovic

This paper is about labeling video frames with action classes under weak supervision in training, where we have access to a temporal ordering of actions, but their start and end frames in training videos are unknown. Following prior work, we use an HMM grounded on a Gated Recurrent Unit (GRU) for frame labeling. Our key contribution is a new constrained discriminative forward loss (CDFL) that we use for training the HMM and GRU under weak supervision. While prior work typically estimates the loss on a single, inferred video segmentation, our CDFL discriminates between the energy of all valid and invalid frame labelings of a training video. A valid frame labeling satisfies the ground-truth temporal ordering of actions, whereas an invalid one violates the ground truth. We specify an efficient recursive algorithm for computing the CDFL in terms of the logadd function of the segmentation energy. Our evaluation on action segmentation and alignment gives superior results to those of the state of the art on the benchmark Breakfast Action, Hollywood Extended, and 50Salads datasets.

Khoi Nguyen, Sinisa Todorovic

This paper is about few-shot segmentation of foreground objects in images. We train a CNN on small subsets of training images, each mimicking the few-shot setting. In each subset, one image serves as the query and the other(s) as support image(s) with ground-truth segmentation. The CNN first extracts feature maps from the query and support images. Then, a class feature vector is computed as an average of the support's feature maps over the known foreground. Finally, the target object is segmented in the query image by using a cosine similarity between the class feature vector and the query's feature map. We make two contributions by: (1) Improving discriminativeness of features so their activations are high on the foreground and low elsewhere; and (2) Boosting inference with an ensemble of experts guided with the gradient of loss incurred when segmenting the support images in testing. Our evaluations on the PASCAL-$5^i$ and COCO-$20^i$ datasets demonstrate that we significantly outperform existing approaches.

Arjun R. Akula, Changsong Liu, Sari Saba-Sadiya et al.

We present a new explainable AI (XAI) framework aimed at increasing justified human trust and reliance in the AI machine through explanations. We pose explanation as an iterative communication process, i.e. dialog, between the machine and human user. More concretely, the machine generates sequence of explanations in a dialog which takes into account three important aspects at each dialog turn: (a) human's intention (or curiosity); (b) human's understanding of the machine; and (c) machine's understanding of the human user. To do this, we use Theory of Mind (ToM) which helps us in explicitly modeling human's intention, machine's mind as inferred by the human as well as human's mind as inferred by the machine. In other words, these explicit mental representations in ToM are incorporated to learn an optimal explanation policy that takes into account human's perception and beliefs. Furthermore, we also show that ToM facilitates in quantitatively measuring justified human trust in the machine by comparing all the three mental representations. We applied our framework to three visual recognition tasks, namely, image classification, action recognition, and human body pose estimation. We argue that our ToM based explanations are practical and more natural for both expert and non-expert users to understand the internal workings of complex machine learning models. To the best of our knowledge, this is the first work to derive explanations using ToM. Extensive human study experiments verify our hypotheses, showing that the proposed explanations significantly outperform the state-of-the-art XAI methods in terms of all the standard quantitative and qualitative XAI evaluation metrics including human trust, reliance, and explanation satisfaction.

Arjun R Akula, Sinisa Todorovic, Joyce Y Chai et al.

This paper presents an explainable AI (XAI) system that provides explanations for its predictions. The system consists of two key components -- namely, the prediction And-Or graph (AOG) model for recognizing and localizing concepts of interest in input data, and the XAI model for providing explanations to the user about the AOG's predictions. In this work, we focus on the XAI model specified to interact with the user in natural language, whereas the AOG's predictions are considered given and represented by the corresponding parse graphs (pg's) of the AOG. Our XAI model takes pg's as input and provides answers to the user's questions using the following types of reasoning: direct evidence (e.g., detection scores), part-based inference (e.g., detected parts provide evidence for the concept asked), and other evidences from spatio-temporal context (e.g., constraints from the spatio-temporal surround). We identify several correlations between user's questions and the XAI answers using Youtube Action dataset.

Tianmin Shu, Sinisa Todorovic, Song-Chun Zhu

This work is about recognizing human activities occurring in videos at distinct semantic levels, including individual actions, interactions, and group activities. The recognition is realized using a two-level hierarchy of Long Short-Term Memory (LSTM) networks, forming a feed-forward deep architecture, which can be trained end-to-end. In comparison with existing architectures of LSTMs, we make two key contributions giving the name to our approach as Confidence-Energy Recurrent Network -- CERN. First, instead of using the common softmax layer for prediction, we specify a novel energy layer (EL) for estimating the energy of our predictions. Second, rather than finding the common minimum-energy class assignment, which may be numerically unstable under uncertainty, we specify that the EL additionally computes the p-values of the solutions, and in this way estimates the most confident energy minimum. The evaluation on the Collective Activity and Volleyball datasets demonstrates: (i) advantages of our two contributions relative to the common softmax and energy-minimization formulations and (ii) a superior performance relative to the state-of-the-art approaches.

Peng Lei, Fuxin Li, Sinisa Todorovic

Using deep learning, this paper addresses the problem of joint object boundary detection and boundary motion estimation in videos, which we named boundary flow estimation. Boundary flow is an important mid-level visual cue as boundaries characterize objects spatial extents, and the flow indicates objects motions and interactions. Yet, most prior work on motion estimation has focused on dense object motion or feature points that may not necessarily reside on boundaries. For boundary flow estimation, we specify a new fully convolutional Siamese network (FCSN) that jointly estimates object-level boundaries in two consecutive frames. Boundary correspondences in the two frames are predicted by the same FCSN with a new, unconventional deconvolution approach. Finally, the boundary flow estimate is improved with an edgelet-based filtering. Evaluation is conducted on three tasks: boundary detection in videos, boundary flow estimation, and optical flow estimation. On boundary detection, we achieve the state-of-the-art performance on the benchmark VSB100 dataset. On boundary flow estimation, we present the first results on the Sintel training dataset. For optical flow estimation, we run the recent approach CPMFlow but on the augmented input with our boundary-flow matches, and achieve significant performance improvement on the Sintel benchmark.

Xu Xu, Sinisa Todorovic

This paper addresses 3D shape recognition. Recent work typically represents a 3D shape as a set of binary variables corresponding to 3D voxels of a uniform 3D grid centered on the shape, and resorts to deep convolutional neural networks(CNNs) for modeling these binary variables. Robust learning of such CNNs is currently limited by the small datasets of 3D shapes available, an order of magnitude smaller than other common datasets in computer vision. Related work typically deals with the small training datasets using a number of ad hoc, hand-tuning strategies. To address this issue, we formulate CNN learning as a beam search aimed at identifying an optimal CNN architecture, namely, the number of layers, nodes, and their connectivity in the network, as well as estimating parameters of such an optimal CNN. Each state of the beam search corresponds to a candidate CNN. Two types of actions are defined to add new convolutional filters or new convolutional layers to a parent CNN, and thus transition to children states. The utility function of each action is efficiently computed by transferring parameter values of the parent CNN to its children, thereby enabling an efficient beam search. Our experimental evaluation on the 3D ModelNet dataset demonstrates that our model pursuit using the beam search yields a CNN with superior performance on 3D shape classification than the state of the art.

Behrooz Mahasseni, Sinisa Todorovic, Alan Fern

This paper formulates and presents a solution to the new problem of budgeted semantic video segmentation. Given a video, the goal is to accurately assign a semantic class label to every pixel in the video within a specified time budget. Typical approaches to such labeling problems, such as Conditional Random Fields (CRFs), focus on maximizing accuracy but do not provide a principled method for satisfying a time budget. For video data, the time required by CRF and related methods is often dominated by the time to compute low-level descriptors of supervoxels across the video. Our key contribution is the new budgeted inference framework for CRF models that intelligently selects the most useful subsets of descriptors to run on subsets of supervoxels within the time budget. The objective is to maintain an accuracy as close as possible to the CRF model with no time bound, while remaining within the time budget. Our second contribution is the algorithm for learning a policy for the sparse selection of supervoxels and their descriptors for budgeted CRF inference. This learning algorithm is derived by casting our problem in the framework of Markov Decision Processes, and then instantiating a state-of-the-art policy learning algorithm known as Classification-Based Approximate Policy Iteration. Our experiments on multiple video datasets show that our learning approach and framework is able to significantly reduce computation time, and maintain competitive accuracy under varying budgets.

Dan Xie, Tianmin Shu, Sinisa Todorovic et al.

This paper is about detecting functional objects and inferring human intentions in surveillance videos of public spaces. People in the videos are expected to intentionally take shortest paths toward functional objects subject to obstacles, where people can satisfy certain needs (e.g., a vending machine can quench thirst), by following one of three possible intent behaviors: reach a single functional object and stop, or sequentially visit several functional objects, or initially start moving toward one goal but then change the intent to move toward another. Since detecting functional objects in low-resolution surveillance videos is typically unreliable, we call them "dark matter" characterized by the functionality to attract people. We formulate the Agent-based Lagrangian Mechanics wherein human trajectories are probabilistically modeled as motions of agents in many layers of "dark-energy" fields, where each agent can select a particular force field to affect its motions, and thus define the minimum-energy Dijkstra path toward the corresponding source "dark matter". For evaluation, we compiled and annotated a new dataset. The results demonstrate our effectiveness in predicting human intent behaviors and trajectories, and localizing functional objects, as well as discovering distinct functional classes of objects by clustering human motion behavior in the vicinity of functional objects.

Shell X. Hu, Christopher K. I. Williams, Sinisa Todorovic

This paper presents a new probabilistic generative model for image segmentation, i.e. the task of partitioning an image into homogeneous regions. Our model is grounded on a mid-level image representation, called a region tree, in which regions are recursively split into subregions until superpixels are reached. Given the region tree, image segmentation is formalized as sampling cuts in the tree from the model. Inference for the cuts is exact, and formulated using dynamic programming. Our tree-cut model can be tuned to sample segmentations at a particular scale of interest out of many possible multiscale image segmentations. This generalizes the common notion that there should be only one correct segmentation per image. Also, it allows moving beyond the standard single-scale evaluation, where the segmentation result for an image is averaged against the corresponding set of coarse and fine human annotations, to conduct a scale-specific evaluation. Our quantitative results are comparable to those of the leading gPb-owt-ucm method, with the notable advantage that we additionally produce a distribution over all possible tree-consistent segmentations of the image.

Tianmin Shu, Dan Xie, Brandon Rothrock et al.

With the advent of drones, aerial video analysis becomes increasingly important; yet, it has received scant attention in the literature. This paper addresses a new problem of parsing low-resolution aerial videos of large spatial areas, in terms of 1) grouping, 2) recognizing events and 3) assigning roles to people engaged in events. We propose a novel framework aimed at conducting joint inference of the above tasks, as reasoning about each in isolation typically fails in our setting. Given noisy tracklets of people and detections of large objects and scene surfaces (e.g., building, grass), we use a spatiotemporal AND-OR graph to drive our joint inference, using Markov Chain Monte Carlo and dynamic programming. We also introduce a new formalism of spatiotemporal templates characterizing latent sub-events. For evaluation, we have collected and released a new aerial videos dataset using a hex-rotor flying over picnic areas rich with group events. Our results demonstrate that we successfully address above inference tasks under challenging conditions.