These are the final papers.
| Date | Class | Meta-Topic | Topic | Presenters | Paper 1 | Paper 2 | Presentation
Review Panels | Things Due |
| Aug 30 2021 | 0 | Getting situated | Intro | fouhey | | | | |
| Sep 1 2021 | 1 | | Reading, writing,
& arithmetic | fouhey, NA, NA | Unsupervised Visual
Representation Learning by
Context Prediction (Doersch et al.
2015) | None | | |
| Sep 6 2021 | 2 | The basics | Labor Day – no class | NA, NA, NA | | | | |
| Sep 8 2021 | 3 | | Architectures | tbd, tbd, tbd | Densely connected convolutional
networks (Huang et al. CVPR 2017) | Exploring Randomly Wired Neural
Networks for Image Recognition
(Xie et al. ICCV 2019) | A; F | |
| Sep 13 2021 | 4 | Getting rid of supervision | Self-supervised learning for pretraining | tbd, tbd, tbd | Momentum Contrast for
Unsupervised Visual
Representation Learning (He et al.
2019) | Exploring Simple Siamese
Representation Learning (Chen et
al. CVPR 2021) | B; G | |
| Sep 15 2021 | 5 | | Reducing supervision for tasks | tbd, tbd, tbd | Space-Time Correspondence as a
Contrastive Random Walk (Jabri et
al. ECCV 2020) | Shelf-Supervised Mesh Prediction
in the Wild (Ye et al. CVPR 2021) | C; H | |
| Sep 20 2021 | 6 | Detection | Detection | tbd, tbd, tbd | MaskRCNN (He et al. ICCV 2017) | PointRend: Image Segmentation as
Rendering (Kirillov et al. CVPR
2020) | D; I | |
| Sep 22 2021 | 7 | | Naming everything | tbd, tbd, tbd | LVIS: A Dataset for Large
Vocabulary Instance Segmentation
(Gupta et al. CVPR 2019) | Evaluating Large-Vocabulary
Object Detectors: The Devil is in
the Details (Dave et al. Arxiv 2021) | E; J | |
| Sep 27 2021 | 8 | Data gone wrong | Dataset bias | tbd, tbd, tbd | An Unbiased Look at Dataset Bias
(Torralba and Efros CVPR 2011) | Do ImageNet Classifiers
Generalize to ImageNet? (Recht et
al. ICML 2019) | F; A | Re-Proposal Due |
| Sep 29 2021 | 9 | | Domain adaptation | tbd, tbd, tbd | Understanding Self-Training for
Gradual Domain Adaptation (Kumar
et al. ICML 2020) | SENTRY: Selective Entropy
Optimization via Committee
Consistency for Unsupervised
Domain Adaptation (Prabhu et al.
ICCV 2021) | G; B | |
| Oct 4 2021 | 10 | What are our models doing | Ethics & fairness | tbd, tbd, tbd | Gender Shades: Intersectional
Accuracy Disparities in
Commercial Gender Classification
(Buolamwini and Gebru FAT 2018) | Dissecting racial bias in an
algorithm used to manage the health
of populations (Obermeyer et al.
Science 2019) | H; C | |
| Oct 6 2021 | 11 | What are we doing? | If I had 100 million dollars | NA, NA, NA | | | I; D | |
| Oct 11 2021 | 12 | 3D | 3D – Implicit functions | tbd, tbd, tbd | PIFu: Pixel-Aligned Implicit
Function for High-Resolution
Clothed Human Digitization (Saito
et al. ICCV 2019) | Occupancy Networks: Learning 3D
Reconstruction in Function Space
(Mescheder et al. CVPR 2019) | J; E | Project Proposal Due |
| Oct 13 2021 | 13 | | 3D – Correspondence and pose | tbd, tbd, tbd | SuperGlue: Learning Feature
Matching with Graph Neural
Networks (Sarlin et al. CVPR 2020) | Bootstrap Your Own
Correspondences (El Banani and
Johnson, ICCV 2021) | A; F | |
| Oct 18 2021 | 14 | Embodiment | NO CLASS
But here are some
nice papers | tbd, tbd, tbd | Semantic Visual Navigation by
Watching Youtube Videos (Chang et
al. NeurIPS 2020) | RMA: Rapid Motor Adaptation for
Legged Robots (Kumar et al., 2021) | B; G | |
| Oct 20 2021 | 15 | | 3D – Multiview | tbd, tbd, tbd | DROID-SLAM: Deep Visual SLAM for
Monocular, Stereo, and RGB-D
Cameras (Teed et al. Arxiv 2021) | SPSG: Self-Supervised
Photometric Scene Generation from
RGB-D Scans (Dai et al. CVPR 2021) | C; H | |
| Oct 25 2021 | 16 | | 3D – Radiance fields | tbd, tbd, tbd | NeRF (Mildenhall et al. ECCV 2020) | pixelNERF (Yu et al. CVPR 2021) | D; I | |
| Oct 27 2021 | 17 | Embodiment | Hands | tbd, tbd, tbd | Reconstructing Hand-Object
Interactions in the Wild (Cao et al.
ICCV 2021) | Towards unconstrained joint
hand-object reconstruction from
RGB videos (Hasson et al. Arxiv
2021) | E; J | |
| Nov 1 2021 | 18 | People | Reconstruction | tbd, tbd, tbd | PARE: Part Attention Regressor for
3D Human Body Estimation (Kocabas
et al. CVPR 2021) | Learning High Fidelity Depths of
Dressed Humans by Watching Social
Media Dance Videos (Jafarian et al.
CVPR 2021) | F; A | |
| Nov 3 2021 | 19 | | Understanding interaction | tbd, tbd, tbd | Rearrangement: A Challenge for
Embodied AI (Batra et al. Arxiv
2020) | Where2Act: From Pixels to Actions
for Articulated 3D Objects (Mo et
al. ICCV 2021) | G; B | |
| Nov 8 2021 | 20 | Transformers and other models | Transformers | tbd, tbd, tbd | Attention Is All You Need (Vaswani
et al. NeurIPS 2017) | An Image is Worth 16x16 Words:
Transformers for Image
Recognition at Scale (Dosovitskiy
et al. Arxiv 2020) | H; C | |
| Nov 10 2021 | 21 | | Others | tbd, tbd, tbd | MLP-Mixer: An all-MLP
Architecture for Vision
(Tolstikhin et al. Arxiv 2021) | Perceiver IO: A General
Architecture for Structured
Inputs & Outputs (Jaegle et al.
Arxiv 2021) | I; D | |
| Nov 15 2021 | 22 | Videos | Video | tbd, tbd, tbd | Rescaling Egocentric Vision
(Damen et al. ECCV 2020) | Ego-Exo: Transferring Visual
Representations from
Third-person to First-person
Videos (Li et al. CVPR 2020) | J; E | |
| Nov 17 2021 | 23 | Science | Science | tbd, tbd, tbd | OpenMonkeyStudio: Automated
Markerless Pose Estimation in
Freely Moving Macaques (Bala et al.
Nature Communications 2020) | Detecting and Tracking Communal
Bird Roosts in Weather Radar Data
(Cheng et al. AAAI 2020) | A; F | Project Preliminary Report Due |
| Nov 22 2021 | 24 | Where are we going? | Where are we going? | NA, NA, NA | | | | |
| Nov 24 2021 | 25 | Thanksgiving | Thanksgiving – no class | NA, NA, NA | | | | |
| Nov 29 2021 | 26 | Other modalities | Language | tbd, tbd, tbd | Learning Transferable Visual
Models From Natural Language
Supervision (Radford et al. 2021) | Less is More: CLIPBERT for
Video-and-Language Learning via
Sparse Sampling (Lei et al. CVPR
2021) | C; H | |
| Dec 1 2021 | 27 | | Sound | tbd, tbd, tbd | Audio-Visual Scene Analysis with
Self-Supervised Multisensory
Features (Owens et al. ECCV 2018) | Self-supervised object detection
from audio-visual correspondence
(Afouras et al. Arxiv 2021) | D; I | |
| Dec 6 2021 | 28 | Generating pictures | Generative models | tbd, tbd, tbd | Generating Diverse High-Fidelity
Images with VQ-VAE-2 (Razavi et al.
NeurIPS 2019) | Zero-Shot Text-to-Image
Generation (Ramesh et al. 2021) | E; J | Project Final Report Draft
|